staging: rtl8723bs: remove commented code block

remove commented 5G code block in hal/hal_com_phycfg.c
which contained removed RT_TRACE log

as reported in the driver's TODO file:

    - find and remove remaining code valid only for 5 GHz.
      Most of the obvious ones have been removed,
      but things like channel > 14 still exist.

so just removed obsolete commented code.

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/b930cfc476c97f2ae5c5b90c1314a6270d56a539.1617640221.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove empty for cycles

remove for-cycles left empty after RT_TRACE deletion
and unused index variables

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/09875f599705d8f50af3511fa6bfaa54b9446a86.1617640221.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: rewrite comparisons to null

fix the following post-commit hook checkpatch issues:

CHECK: Comparison to NULL could be written "!pwep"
147: FILE: drivers/staging/rtl8723bs/os_dep/ioctl_linux.c:442:
+ if (pwep == NULL)

CHECK: Comparison to NULL could be written "!skb"
226: FILE: drivers/staging/rtl8723bs/os_dep/recv_linux.c:204:
+ if (skb == NULL)

CHECK: Comparison to NULL could be written "!dvobj"
275: FILE: drivers/staging/rtl8723bs/os_dep/sdio_intf.c:398:
+ if (dvobj == NULL)

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/92edc9da10838dd99f96a99121343f4ec11b5d90.1617640221.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: place constant on the right side of the test

fix following post-hook checkpatch issue:

WARNING: Comparisons should place the constant on the right side of the test
85: FILE: drivers/staging/rtl8723bs/hal/sdio_halinit.c:676:
+ if (_FAIL == ret)

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/8a6967abaef8b2edad6dc829dd232a7f048899d2.1617640221.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove unnecessary bracks after RT_TRACE deletion

Remove all unnecessary bracks in if blocks, after RT_TRACE macro
deletion

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/19fda6f0a2b381399623f5a5def39b07be0d995f.1617640221.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove empty #ifdef blocks after RT_TRACE deletion

Remove all empty #ifdef blocks left empty after
RT_TRACE macro deletion.

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/f054fa97ff8197bd9d0320a601428c132de95e1b.1617640221.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove empty if, else blocks after RT_TRACE deletion

Remove all if, else if, else blocks left empty after
RT_TRACE macro deletion.

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/0ce82ac515ba5f6b6318e63d910ad61d7a61de35.1617640221.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove RT_TRACE log definitions

remove RT_TRACE log definitions.

Remove all of the RT_TRACE logs in hal/ and os_dep/ file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/163c0f17c88dc8977103a0971c50f4769ad78173.1617640221.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove commented out RT_TRACE logs in hal/ and os_dep/

Remove all commented out TR_TRACE calls.

Remove all of the RT_TRACE logs in hal/ and os_dep/ file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/85338f2b2cffb530efbd987064d146fc2be49c72.1617640221.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove all RT_TRACE logs in hal/ and os_dep/

Remove all of the RT_TRACE logs in hal/ and os_dep/ files as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

This bulk remove has been done with the following semantic
patch:

@@
expression a, b, c;
@@

- RT_TRACE(a, b, (c));

Suggested-by: Joe Perches <joe@perches.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/f5c6f61461ad957ecd5998019ac1ee1215dde097.1617640221.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: bcm2835-pcm: Allow up to 8 channels and 192kHz data rate

The firmware driver can support this, so allow it to be selected
Increase the buffer sizes to handle the higher data rates.

Reviewed-by: Nicolas Saenz Julienne <nsaenz@kernel.org>
Signed-off-by: Dom Cobley <popcornmix@gmail.com>
Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com>
Link: https://lore.kernel.org/r/1617373688-8715-1-git-send-email-stefan.wahren@i2se.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: add spaces around operators in core/rtw_ieee80211.c

fix post-commit hook checkpatch issues:

CHECK: spaces preferred around that '+' (ctx:VxV)
161: FILE: drivers/staging/rtl8723bs/core/rtw_ieee80211.c:648:
+ *wpa_len = in_ie[cnt+1]+2;
                      ^

CHECK: spaces preferred around that '+' (ctx:VxV)
161: FILE: drivers/staging/rtl8723bs/core/rtw_ieee80211.c:648:
+ *wpa_len = in_ie[cnt+1]+2;
                         ^

CHECK: spaces preferred around that '+' (ctx:VxV)
162: FILE: drivers/staging/rtl8723bs/core/rtw_ieee80211.c:649:
+ cnt += in_ie[cnt+1]+2;  /* get next */
                  ^

CHECK: spaces preferred around that '+' (ctx:VxV)
162: FILE: drivers/staging/rtl8723bs/core/rtw_ieee80211.c:649:
+ cnt += in_ie[cnt+1]+2;  /* get next */
                     ^

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/1b5ee3d974c336e20f034d5de449fc29967a6213.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove RT_TRACE logs in core/rtw_ieee80211.c

Remove all of the RT_TRACE logs in the core/rtw_ieee80211.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/f9985fe00edc78b06682aaa9e1a4cb0bdcbbd384.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove RT_TRACE logs in core/rtw_sta_mgt.c

Remove all of the RT_TRACE logs in the core/rtw_sta_mgt.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/155cff4252d13a4eefe12a397e1623fb9ab46f15.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove all RT_TRACE logs in core/rtw_wlan_util.c

Remove all of the RT_TRACE logs in the core/rtw_wlan_util.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/406adfd834b55ea5694dfb3c33c36fe0754edbb3.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: place constant on the right side of the test in core/rtw_ioctl_set.c

fix posst-commit hook checkpatch issues:

WARNING: Comparisons should place the constant on the right
side of the test
40: FILE: drivers/staging/rtl8723bs/core/rtw_ioctl_set.c:71:
+ if (_SUCCESS != ret)

WARNING: Comparisons should place the constant on the right
side of the test
69: FILE: drivers/staging/rtl8723bs/core/rtw_ioctl_set.c:122:
+ if (_SUCCESS != ret)

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/3c701cacd71a6bb0f59242fc3a987f72dabb93e2.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove RT_TRACE logs in core/rtw_ioctl_set.c

Remove all of the RT_TRACE logs in the core/rtw_ioctl_set.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/70a88973cd1475275a725522930dfb575739908b.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove commented RT_TRACE call in core/rtw_ioctl_set.c

Remove commented RT_TRACE call in core/rtw_ioctl_set.c

Remove all of the RT_TRACE logs in the core/rtw_ioctl_set.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/aae1227b56fb75c0b221a46b612aa4aeb6460feb.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: fix comparison in if condition in core/rtw_recv.c

fix post-commit checkpatch issue:

CHECK: Using comparison to false is error prone
27: FILE: drivers/staging/rtl8723bs/core/rtw_recv.c:381:
+ if (psecuritypriv->
bcheck_grpkey == false &&

and fix the same issue in second comparison to true
inside the same if condition
IS_MCAST(prxattrib->ra) == true
^^^^^^^

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/4b363edbc2da36cb3c63edbcd43e3285754768c9.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove unnecessary parentheses in core/rtw_recv.c

fix post-commit checkpatch issue:

CHECK: Unnecessary parentheses around
'psecuritypriv->bcheck_grpkey == false'
24: FILE: drivers/staging/rtl8723bs/core/rtw_recv.c:381:
+ if ((psecuritypriv->
bcheck_grpkey == false) &&
+     (IS_MCAST(prxattrib->ra)
 == true))

and fix same issue on second comparison
(IS_MCAST(prxattrib->ra) == true)
^                               ^

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/eab39c58913d799dc90efb8b234c2a7fdb61ba57.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: split long line in core/rtw_recv.c

fix post-commit hook checkpatch issue:

WARNING: line length of 113 exceeds 100 columns
110: FILE: drivers/staging/rtl8723bs/core/rtw_recv.c:381:
+ if ((psecuritypriv->bcheck_grpkey
 == false) && (IS_MCAST(prxattrib->ra) == true))

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/ac256752eddba7678e8305cb5e560969dec482de.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: added spaces around operator in core/rtw_recv.c

fix post-commit hook checkpatch issue:

CHECK: spaces preferred around that '+' (ctx:VxV)
60: FILE: drivers/staging/rtl8723bs/core/rtw_recv.c:359:
+ if (miccode[i] != *(pframemic+i))
                               ^

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/6202df1b5383aa0a66810358495c2b1c9dd6c189.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove RT_TRACE logs in core/rtw_recv.c

Remove all of the RT_TRACE logs in the core/rtw_recv.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/c4a5d7cc3a1e7c5a13622b277ee34042c34981bd.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove commented RT_TRACE calls in core/rtw_recv.c

Remove commented RT_TRACE calls

Remove all of the RT_TRACE logs in the core/rtw_recv.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/6986ce9c7bb0e9560d9463281973e1858347a966.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove RT_TRACE logs in core/rtw_mlme_ext.c

Remove all of the RT_TRACE logs in the core/rtw_mlme_ext.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/4f838d1ce3cd16fab8f05d1e0f80c2e7469701b8.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: tidy up some error handling in core/rtw_mlme.c

the RT_TRACE() output is not useful so we want to delete it. In this case
there is no cleanup for rtw_cleanbss_cmd() required or even possible. I've
deleted the RT_TRACE() output and added a goto unlock to show
that we can't continue if rtw_createbss_cmd() fails.

Suggested-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/6affc9dd905f7c94a93bbd0fa706bc9026103c2e.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove RT_TRACE logs in core/rtw_mlme.c

Remove all of the RT_TRACE logs in the core/rtw_mlme.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/19cc25a51af207e68a95e76ce384a3f7b7d2c408.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove commented RT_TRACE calls in core/rtw_mlme.c

Remove commented RT_TRACE calls in core/rtw_mlme.c

Remove all of the RT_TRACE logs in the core/rtw_mlme.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/09862e8783366045a08fd97523595710a08df918.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove unnecessary parentheses in if condition in core/rtw_cmd.c

fix post-commit hook checkpatch issue:

WARNING: Unnecessary parentheses
166: FILE: drivers/staging/rtl8723bs/core/rtw_cmd.c:2000:
+ if ((pcmd->res != H2C_SUCCESS))

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/04aa9d68d6d28d295001dde7cc14f33ef83f8046.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: fix null check conditions in core/rtw_cmd.c

fix post-commit hook checkpatch issues:

CHECK: Comparison to NULL could be written "!psta"
178: FILE: drivers/staging/rtl8723bs/core/rtw_cmd.c:2012:
+ if (psta == NULL)

CHECK: Comparison to NULL could be written "!psta"
198: FILE: drivers/staging/rtl8723bs/core/rtw_cmd.c:2066:
+ if (psta == NULL)

CHECK: Comparison to NULL could be written "!psta"
211: FILE: drivers/staging/rtl8723bs/core/rtw_cmd.c:2081:
+ if (psta == NULL)

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/03e36e1710ea4dfd07c3753d9a8df5746e59bc18.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove RT_TRACE logs in core/rtw_cmd.c

Remove all of the RT_TRACE logs in the core/rtw_cmd.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

removed a variable, left unused after RT_TRACE removal

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/6480425f7b4211ff1fe0bad0e8062d63c4ed9af3.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove unnecessary parentheses in if-condition in core/rtw_pwrctrl.c

fix following post-commit hook checkpatch issues:

CHECK: Unnecessary parentheses around 'pwrpriv->rpwm == pslv'
26: FILE: drivers/staging/rtl8723bs/core/rtw_pwrctrl.c:269:
+ if ((pwrpriv->rpwm == pslv) ||
+     ((pwrpriv->rpwm >= PS_STATE_S2)
&& (pslv >= PS_STATE_S2)))

CHECK: Unnecessary parentheses around 'pwrpriv->rpwm >= PS_STATE_S2'
26: FILE: drivers/staging/rtl8723bs/core/rtw_pwrctrl.c:269:
+ if ((pwrpriv->rpwm == pslv) ||
+     ((pwrpriv->rpwm >= PS_STATE_S2)
&& (pslv >= PS_STATE_S2)))

CHECK: Unnecessary parentheses around 'pslv >= PS_STATE_S2'
26: FILE: drivers/staging/rtl8723bs/core/rtw_pwrctrl.c:269:
+ if ((pwrpriv->rpwm == pslv) ||
+     ((pwrpriv->rpwm >= PS_STATE_S2)
&& (pslv >= PS_STATE_S2)))

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/d4d1d256586cbdfc55157b0b384b28d53a61c2b8.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: fix logical continuation issue in core/rtw_pwrctrl.c

fix following post-commit hook checkpatch issue:

CHECK: Logical continuations should be on the previous line
22: FILE: drivers/staging/rtl8723bs/core/rtw_pwrctrl.c:270:
  if ((pwrpriv->rpwm == pslv)
+ || ((pwrpriv->rpwm >= PS_STATE_S2)
 && (pslv >= PS_STATE_S2)))

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/ecbd557c645e280a801c34d235559d2d138ec67f.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove all RT_TRACE logs in core/rtw_pwrctrl.c

Remove all of the RT_TRACE logs in the core/rtw_pwrctrl.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/a842ef8f93c920e52fed1b552968b679ca5e7be6.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: fix error prone if conditions in core/rtw_eeprom.c

fix the following post-commit hook checkpatch issues:

CHECK: Using comparison to true is error prone
21: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:38:
+ if (padapter->bSurpriseRemoved == true)

CHECK: Using comparison to true is error prone
36: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:50:
+ if (padapter->bSurpriseRemoved == true)

CHECK: Using comparison to true is error prone
47: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:59:
+ if (padapter->bSurpriseRemoved == true)

CHECK: Using comparison to true is error prone
60: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:72:
+ if (padapter->bSurpriseRemoved == true)

CHECK: Using comparison to true is error prone
73: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:83:
+ if (padapter->bSurpriseRemoved == true)

CHECK: Using comparison to true is error prone
86: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:120:
+ if (padapter->bSurpriseRemoved == true)

CHECK: Using comparison to true is error prone
93: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:124:
+ if (padapter->bSurpriseRemoved == true)

CHECK: Using comparison to true is error prone
101: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:129:
+ if (padapter->bSurpriseRemoved == true)

CHECK: Using comparison to true is error prone
108: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:133:
+ if (padapter->bSurpriseRemoved == true)

CHECK: Using comparison to true is error prone
121: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:149:
+ if (padapter->bSurpriseRemoved == true)

CHECK: Using comparison to true is error prone
130: FILE: drivers/staging/rtl8723bs/core/rtw_eeprom.c:155:
+ if (padapter->bSurpriseRemoved == true)

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/2ffcefe0b2c44cb5d392c4cd095da0f9b754cdc7.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove all RT_TRACE logs in core/rtw_eeprom.c

Remove all of the RT_TRACE logs in the core/rtw_eeprom.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/e302fadf18a36d8521456858caff641c0f598731.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: fix spaces around operator issues in core/rtw_security.c

fix the following post commit hook checkpatch issues:

CHECK: spaces preferred around that '-' (ctx:VxV)
25: FILE: drivers/staging/rtl8723bs/core/rtw_security.c:729:
+ if (crc[3] != payload[length-1] ||
crc[2] != payload[length-2] ||
                              ^

CHECK: spaces preferred around that '-' (ctx:VxV)
25: FILE: drivers/staging/rtl8723bs/core/rtw_security.c:729:
+ if (crc[3] != payload[length-1] ||
crc[2] != payload[length-2] ||
                                ^

CHECK: spaces preferred around that '-' (ctx:VxV)
26: FILE: drivers/staging/rtl8723bs/core/rtw_security.c:730:
+     crc[1] != payload[length-3] ||
crc[0] != payload[length-4])
                              ^

CHECK: spaces preferred around that '-' (ctx:VxV)
26: FILE: drivers/staging/rtl8723bs/core/rtw_security.c:730:
+     crc[1] != payload[length-3] ||
crc[0] != payload[length-4])
^

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/9143a02ae6959d4deae6c22bdba42041bf6484a2.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: fix line exceed warning in core/rtw_security.c

fix the following post-commit hook checkpatch issue:

WARNING: line length of 149 exceeds 100 columns
61: FILE: drivers/staging/rtl8723bs/core/rtw_security.c:729:
+ if (crc[3] != payload[length-1] ||
crc[2] != payload[length-2] || crc[1] != payload[length-3] ||
crc[0] != payload[length-4])

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/862862a3e309209e2508de0c9a1b5a02ce1312b2.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove RT_TRACE logs in core/rtw_security.c

Remove all of the RT_TRACE logs in the core/rtw_security.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/14207c2b7ea789e8b0f3bd37d2758e979eff997c.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: fix condition in if statement in core/rtw_xmit.c

fix post-commit hook checkpatch issue:

CHECK: Using comparison to true is error prone
71: FILE: drivers/staging/rtl8723bs/core/rtw_xmit.c:565:
+ ((padapter->securitypriv.sw_encrypt == true) ||
 (psecuritypriv->hw_decrypted == false)))

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/45793b555f46312ea3c9cd285886dec22f3a1131.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove RT_TRACE logs in core/rtw_xmit.c

Remove all of the RT_TRACE logs in the core/rtw_xmit.c file as they
currently do nothing as they require the code to be modified by
hand in order to be turned on. This obviously has not happened
since the code was merged. Moreover it relies on an unneeded
private log level tracing which overrides the in-kernel public one,
so just remove them as they are unused.

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/c7e5ba68189e4e7ca923741d31908dda7850e45e.1617545239.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: fbtft: convert sysfs snprintf to sysfs_emit

Fix the following coccicheck warning:
drivers/staging/fbtft//fbtft-sysfs.c:202:8-16:
WARNING: use scnprintf or sprintf

Signed-off-by: Xuezhi Zhang <zhangxuezhi1@yulong.com>
Link: https://lore.kernel.org/r/20210403140722.169623-1-llyz108@163.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: iio: cdc: remove repeated word

Fix checkpatch warning:
WARNING: Possible repeated word: 'from'

Signed-off-by: David Villasana Jiménez <davidvillasana14@gmail.com>
Link: https://lore.kernel.org/r/YGo8q4R5dqgtDMlg@fedora
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8192e: move const after static

Move the const after static to conform with kernel preference for
static const <type> declaration style instead of static <type> const.
Identified by checkpatch: WARNING: Move const after static - use 'static
const char'.

Signed-off-by: Deborah Brouwer <deborahbrouwer3563@gmail.com>
Link: https://lore.kernel.org/r/20210404222349.7218-1-deborahbrouwer3563@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8712: remove extra blank lines

Remove extra blank lines to adhere to Linux kernel coding style.
Reported by checkpatch.

Signed-off-by: Zhansaya Bagdauletkyzy <zhansayabagdaulet@gmail.com>
Link: https://lore.kernel.org/r/d56183614a04590bdf3a6280b7f23664a5bb394c.1617568354.git.zhansayabagdaulet@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8712: add blank lines after declarations

Add blank lines after function/struct/union/enum declarations to adhere to
Linux kernel coding style.
Reported by checkpatch.

Signed-off-by: Zhansaya Bagdauletkyzy <zhansayabagdaulet@gmail.com>
Link: https://lore.kernel.org/r/ffe2ab70ef3cb73d3d6dd19d88804af7ecb568a2.1617568354.git.zhansayabagdaulet@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: sm750fb: Remove unnecessary blank line

Remove unnecessary extra blank line to improve readability.
Reported by checkpatch.

Signed-off-by: Zhansaya Bagdauletkyzy <zhansayabagdaulet@gmail.com>
Link: https://lore.kernel.org/r/20210404111055.GA22626@zhans
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8188eu: core: replace spaces with a tab

To conform to Linux kernel coding style, a tab replaces the two spaces
that were previously used to indent brackets.  Identified by the
checkpatch warning: WARNING: please, no spaces at the start of a line.

Signed-off-by: Deborah Brouwer <deborahbrouwer3563@gmail.com>
Link: https://lore.kernel.org/r/20210403224738.28957-1-deborahbrouwer3563@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8188eu: core: add comma within a comment

Add a comma to separate repeated words in a comment. The comma preserves
the meaning of the comment while also stopping the checkpatch warning:
WARNING: Possible repeated word: 'very'.

Signed-off-by: Deborah Brouwer <deborahbrouwer3563@gmail.com>
Link: https://lore.kernel.org/r/20210403210930.17781-1-deborahbrouwer3563@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge 5.12-rc6 into staging-next

We need the staging fixes in here as well.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Linux 5.12-rc6

firewire: nosy: Fix a use-after-free bug in nosy_ioctl()

For each device, the nosy driver allocates a pcilynx structure.
A use-after-free might happen in the following scenario:

 1. Open nosy device for the first time and call ioctl with command
    NOSY_IOC_START, then a new client A will be malloced and added to
    doubly linked list.
 2. Open nosy device for the second time and call ioctl with command
    NOSY_IOC_START, then a new client B will be malloced and added to
    doubly linked list.
 3. Call ioctl with command NOSY_IOC_START for client A, then client A
    will be readded to the doubly linked list. Now the doubly linked
    list is messed up.
 4. Close the first nosy device and nosy_release will be called. In
    nosy_release, client A will be unlinked and freed.
 5. Close the second nosy device, and client A will be referenced,
    resulting in UAF.

The root cause of this bug is that the element in the doubly linked list
is reentered into the list.

Fix this bug by adding a check before inserting a client.  If a client
is already in the linked list, don't insert it.

The following KASAN report reveals it:

   BUG: KASAN: use-after-free in nosy_release+0x1ea/0x210
   Write of size 8 at addr ffff888102ad7360 by task poc
   CPU: 3 PID: 337 Comm: poc Not tainted 5.12.0-rc5+ #6
   Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.0-59-gc9ba5276e321-prebuilt.qemu.org 04/01/2014
   Call Trace:
     nosy_release+0x1ea/0x210
     __fput+0x1e2/0x840
     task_work_run+0xe8/0x180
     exit_to_user_mode_prepare+0x114/0x120
     syscall_exit_to_user_mode+0x1d/0x40
     entry_SYSCALL_64_after_hwframe+0x44/0xae

   Allocated by task 337:
     nosy_open+0x154/0x4d0
     misc_open+0x2ec/0x410
     chrdev_open+0x20d/0x5a0
     do_dentry_open+0x40f/0xe80
     path_openat+0x1cf9/0x37b0
     do_filp_open+0x16d/0x390
     do_sys_openat2+0x11d/0x360
     __x64_sys_open+0xfd/0x1a0
     do_syscall_64+0x33/0x40
     entry_SYSCALL_64_after_hwframe+0x44/0xae

   Freed by task 337:
     kfree+0x8f/0x210
     nosy_release+0x158/0x210
     __fput+0x1e2/0x840
     task_work_run+0xe8/0x180
     exit_to_user_mode_prepare+0x114/0x120
     syscall_exit_to_user_mode+0x1d/0x40
     entry_SYSCALL_64_after_hwframe+0x44/0xae

   The buggy address belongs to the object at ffff888102ad7300 which belongs to the cache kmalloc-128 of size 128
   The buggy address is located 96 bytes inside of 128-byte region [ffff888102ad7300, ffff888102ad7380)

[ Modified to use 'list_empty()' inside proper lock  - Linus ]

Link: https://lore.kernel.org/lkml/1617433116-5930-1-git-send-email-zheyuma97@gmail.com/
Reported-and-tested-by: 马哲宇 (Zheyu Ma) <zheyuma97@gmail.com>
Signed-off-by: Zheyu Ma <zheyuma97@gmail.com>
Cc: Greg Kroah-Hartman <greg@kroah.com>
Cc: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Merge tag 'for-linus' of git://github.com/openrisc/linux

Pull OpenRISC fix from Stafford Horne:
 "Fix duplicate header include in Litex SOC driver"

* tag 'for-linus' of git://github.com/openrisc/linux:
  soc: litex: Remove duplicated header file inclusion

Merge tag 'io_uring-5.12-2021-04-03' of git://git.kernel.dk/linux-block

POull io_uring fix from Jens Axboe:
 "Just fixing a silly braino in a previous patch, where we'd end up
  failing to compile if CONFIG_BLOCK isn't enabled.

  Not that a lot of people do that, but kernel bot spotted it and it's
  probably prudent to just flush this out now before -rc6.

  Sorry about that, none of my test compile configs have !CONFIG_BLOCK"

* tag 'io_uring-5.12-2021-04-03' of git://git.kernel.dk/linux-block:
  io_uring: fix !CONFIG_BLOCK compilation failure

soc: litex: Remove duplicated header file inclusion

The header file <linux/errno.h> is already included above and can be
removed here.

Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Signed-off-by: Mateusz Holenko <mholenko@antmicro.com>
Signed-off-by: Stafford Horne <shorne@gmail.com>

Merge tag 'gfs2-v5.12-rc2-fixes2' of git://git./linux/kernel/git/gfs2/linux-gfs2

Pull gfs2 fixes from Andreas Gruenbacher:
 "Two more gfs2 fixes"

* tag 'gfs2-v5.12-rc2-fixes2' of git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2:
  gfs2: report "already frozen/thawed" errors
  gfs2: Flag a withdraw if init_threads() fails

Merge tag 'riscv-for-linus-5.12-rc6' of git://git./linux/kernel/git/riscv/linux

Pull RISC-V fixes from Palmer Dabbelt:
 "A handful of fixes for 5.12:

   - fix a stack tracing regression related to "const register asm"
     variables, which have unexpected behavior.

   - ensure the value to be written by put_user() is evaluated before
     enabling access to userspace memory..

   - align the exception vector table correctly, so we don't rely on the
     firmware's handling of unaligned accesses.

   - build fix to make NUMA depend on MMU, which triggered on some
     randconfigs"

* tag 'riscv-for-linus-5.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
  riscv: Make NUMA depend on MMU
  riscv: remove unneeded semicolon
  riscv,entry: fix misaligned base for excp_vect_table
  riscv: evaluate put_user() arg before enabling user access
  riscv: Drop const annotation for sp

Merge tag 'powerpc-5.12-5' of git://git./linux/kernel/git/powerpc/linux

Pull powerpc fixes from Michael Ellerman:
 "Fix a bug on pseries where spurious wakeups from H_PROD would prevent
  partition migration from succeeding.

  Fix oopses seen in pcpu_alloc(), caused by parallel faults of the
  percpu mapping causing us to corrupt the protection key used for the
  mapping, and cause a fatal key fault.

  Thanks to Aneesh Kumar K.V, Murilo Opsfelder Araujo, and Nathan Lynch"

* tag 'powerpc-5.12-5' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc/mm/book3s64: Use the correct storage key value when calling H_PROTECT
  powerpc/pseries/mobility: handle premature return from H_JOIN
  powerpc/pseries/mobility: use struct for shared state

Merge tag 'hyperv-fixes-signed-20210402' of git://git./linux/kernel/git/hyperv/linux

Pull Hyper-V fixes from Wei Liu:
 "One fix from Lu Yunlong for a double free in hvfb_probe"

* tag 'hyperv-fixes-signed-20210402' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux:
  video: hyperv_fb: Fix a double free in hvfb_probe

Merge tag 'driver-core-5.12-rc6' of git://git./linux/kernel/git/gregkh/driver-core

Pull driver core fix from Greg KH:
 "Here is a single driver core fix for a reported problem with differed
  probing. It has been in linux-next for a while with no reported
  problems"

* tag 'driver-core-5.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
  driver core: clear deferred probe reason on probe retry

Merge tag 'char-misc-5.12-rc6' of git://git./linux/kernel/git/gregkh/char-misc

Pull char/misc driver fixes from Greg KH:
 "Here are a few small driver char/misc changes for 5.12-rc6.

  Nothing major here, a few fixes for reported issues:

   - interconnect fixes for problems found

   - fbcon syzbot-found fix

   - extcon fixes

   - firmware stratix10 bugfix

   - MAINTAINERS file update.

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'char-misc-5.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  drivers: video: fbcon: fix NULL dereference in fbcon_cursor()
  mei: allow map and unmap of client dma buffer only for disconnected client
  MAINTAINERS: Add linux-phy list and patchwork
  interconnect: Fix kerneldoc warning
  firmware: stratix10-svc: reset COMMAND_RECONFIG_FLAG_PARTIAL to 0
  extcon: Fix error handling in extcon_dev_register
  extcon: Add stubs for extcon_register_notifier_all() functions
  interconnect: core: fix error return code of icc_link_destroy()
  interconnect: qcom: msm8939: remove rpm-ids from non-RPM nodes

Merge tag 'staging-5.12-rc6' of git://git./linux/kernel/git/gregkh/staging

Pull staging driver fixes from Greg KH:
 "Here are two rtl8192e staging driver fixes for reported problems.

  Both of these have been in linux-next for a while with no reported
  issues"

* tag 'staging-5.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  staging: rtl8192e: Change state information from u16 to u8
  staging: rtl8192e: Fix incorrect source in memcpy()

Merge tag 'tty-5.12-rc6' of git://git./linux/kernel/git/gregkh/tty

Pull serial driver fix from Greg KH:
 "Here is a single serial driver fix for 5.12-rc6. Is is a revert of a
  change that showed up in 5.9 that has been reported to cause problems.

  It has been in linux-next for a while with no reported issues"

* tag 'tty-5.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  soc: qcom-geni-se: Cleanup the code to remove proxy votes

Merge tag 'usb-5.12-rc6' of git://git./linux/kernel/git/gregkh/usb

Pull USB fixes from Greg KH:
 "Here are a few small USB driver fixes for 5.12-rc6 to resolve reported
  problems.

  They include:

   - a number of cdc-acm fixes for reported problems. It seems more
     people are using this driver lately...

   - dwc3 driver fixes for reported problems, and fixes for the fixes :)

   - dwc2 driver fixes for reported issues.

   - musb driver fix.

   - new USB quirk additions.

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'usb-5.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (23 commits)
  usb: dwc2: Prevent core suspend when port connection flag is 0
  usb: dwc2: Fix HPRT0.PrtSusp bit setting for HiKey 960 board.
  usb: musb: Fix suspend with devices connected for a64
  usb: xhci-mtk: fix broken streams issue on 0.96 xHCI
  usb: dwc3: gadget: Clear DEP flags after stop transfers in ep disable
  usbip: vhci_hcd fix shift out-of-bounds in vhci_hub_control()
  USB: quirks: ignore remote wake-up on Fibocom L850-GL LTE modem
  USB: cdc-acm: do not log successful probe on later errors
  USB: cdc-acm: always claim data interface
  USB: cdc-acm: use negation for NULL checks
  USB: cdc-acm: clean up probe error labels
  USB: cdc-acm: drop redundant driver-data reset
  USB: cdc-acm: drop redundant driver-data assignment
  USB: cdc-acm: fix use-after-free after probe failure
  USB: cdc-acm: fix double free on probe failure
  USB: cdc-acm: downgrade message to debug
  USB: cdc-acm: untangle a circular dependency between callback and softint
  cdc-acm: fix BREAK rx code path adding necessary calls
  usb: gadget: udc: amd5536udc_pci fix null-ptr-dereference
  usb: dwc3: pci: Enable dis_uX_susphy_quirk for Intel Merrifield
  ...

Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fix from James Bottomley:
 "A single fix to iscsi for a rare race condition which can cause a
  kernel panic"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: iscsi: Fix race condition between login and sync thread

io_uring: fix !CONFIG_BLOCK compilation failure

kernel test robot correctly pinpoints a compilation failure if
CONFIG_BLOCK isn't set:

fs/io_uring.c: In function '__io_complete_rw':
>> fs/io_uring.c:2509:48: error: implicit declaration of function 'io_rw_should_reissue'; did you mean 'io_rw_reissue'? [-Werror=implicit-function-declaration]
    2509 |  if ((res == -EAGAIN || res == -EOPNOTSUPP) && io_rw_should_reissue(req)) {
         |                                                ^~~~~~~~~~~~~~~~~~~~
         |                                                io_rw_reissue
    cc1: some warnings being treated as errors

Ensure that we have a stub declaration of io_rw_should_reissue() for
!CONFIG_BLOCK.

Fixes: 230d50d448ac ("io_uring: move reissue into regular IO path")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Merge tag 'block-5.12-2021-04-02' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:

 - Remove comment that never came to fruition in 22 years of development
   (Christoph)

 - Remove unused request flag (Christoph)

 - Fix for null_blk fake timeout handling (Damien)

 - Fix for IOCB_NOWAIT being ignored for O_DIRECT on raw bdevs (Pavel)

 - Error propagation fix for multiple split bios (Yufen)

* tag 'block-5.12-2021-04-02' of git://git.kernel.dk/linux-block:
  block: remove the unused RQF_ALLOCED flag
  block: update a few comments in uapi/linux/blkpg.h
  block: don't ignore REQ_NOWAIT for direct IO
  null_blk: fix command timeout completion handling
  block: only update parent bi_status when bio fail

Merge tag 'io_uring-5.12-2021-04-02' of git://git.kernel.dk/linux-block

Pull io_uring fixes from Jens Axboe:
 "Nothing really major in here, and finally nothing really related to
  signals. A few minor fixups related to the threading changes, and some
  general fixes, that's it.

  There's the pending gdb-get-confused-about-arch, but that's more of a
  cosmetic issue, nothing that hinder use of it. And given that other
  archs will likely be affected by that oddity too, better to postpone
  any changes there until 5.13 imho"

* tag 'io_uring-5.12-2021-04-02' of git://git.kernel.dk/linux-block:
  io_uring: move reissue into regular IO path
  io_uring: fix EIOCBQUEUED iter revert
  io_uring/io-wq: protect against sprintf overflow
  io_uring: don't mark S_ISBLK async work as unbounded
  io_uring: drop sqd lock before handling signals for SQPOLL
  io_uring: handle setup-failed ctx in kill_timeouts
  io_uring: always go for cancellation spin on exec

Merge tag 'acpi-5.12-rc6' of git://git./linux/kernel/git/rafael/linux-pm

Pull ACPI fixes from Rafael Wysocki:
 "These fix an ACPI tables management issue, an issue related to the
  ACPI enumeration of devices and CPU wakeup in the ACPI processor
  driver.

  Specifics:

   - Ensure that the memory occupied by ACPI tables on x86 will always
     be reserved to prevent it from being allocated for other purposes
     which was possible in some cases (Rafael Wysocki).

   - Fix the ACPI device enumeration code to prevent it from attempting
     to evaluate the _STA control method for devices with unmet
     dependencies which is likely to fail (Hans de Goede).

   - Fix the handling of CPU0 wakeup in the ACPI processor driver to
     prevent CPU0 online failures from occurring (Vitaly Kuznetsov)"

* tag 'acpi-5.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: processor: Fix CPU0 wakeup in acpi_idle_play_dead()
  ACPI: scan: Fix _STA getting called on devices with unmet dependencies
  ACPI: tables: x86: Reserve memory occupied by ACPI tables

Merge tag 'pm-5.12-rc6' of git://git./linux/kernel/git/rafael/linux-pm

Pull power management fixes from Rafael Wysocki:
 "These fix a race condition and an ordering issue related to using
  device links in the runtime PM framework and two kerneldoc comments in
  cpufreq.

  Specifics:

   - Fix race condition related to the handling of supplier devices
     during consumer device probe and fix the order of decrementation of
     two related reference counters in the runtime PM core code handling
     supplier devices (Adrian Hunter).

   - Fix kerneldoc comments in cpufreq that have not been updated along
     with the functions documented by them (Geert Uytterhoeven)"

* tag 'pm-5.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  PM: runtime: Fix race getting/putting suppliers at probe
  PM: runtime: Fix ordering in pm_runtime_get_suppliers()
  cpufreq: Fix scaling_{available,boost}_frequencies_show() comments

block: remove the unused RQF_ALLOCED flag

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block: update a few comments in uapi/linux/blkpg.h

The big top of the file comment talk about grand plans that never
happened, so remove them to not confuse the readers.  Also mark the
devname and volname fields as ignored as they were never used by the
kernel.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

Merge tag 'trace-v5.12-rc5-2' of git://git./linux/kernel/git/rostedt/linux-trace

Pull tracing fix from Steven Rostedt:
 "Fix stack trace entry size to stop showing garbage

  The macro that creates both the structure and the format displayed to
  user space for the stack trace event was changed a while ago to fix
  the parsing by user space tooling. But this change also modified the
  structure used to store the stack trace event. It changed the caller
  array field from [0] to [8].

  Even though the size in the ring buffer is dynamic and can be
  something other than 8 (user space knows how to handle this), the 8
  extra words was not accounted for when reserving the event on the ring
  buffer, and added 8 more entries, due to the calculation of
  "sizeof(*entry) + nr_entries * sizeof(long)", as the sizeof(*entry)
  now contains 8 entries.

  The size of the caller field needs to be subtracted from the size of
  the entry to create the correct allocation size"

* tag 'trace-v5.12-rc5-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing: Fix stack trace event size

io_uring: move reissue into regular IO path

It's non-obvious how retry is done for block backed files, when it happens
off the kiocb done path. It also makes it tricky to deal with the iov_iter
handling.

Just mark the req as needing a reissue, and handling it from the
submission path instead. This makes it directly obvious that we're not
re-importing the iovec from userspace past the submit point, and it means
that we can just reuse our usual -EAGAIN retry path from the read/write
handling.

At some point in the future, we'll gain the ability to always reliably
return -EAGAIN through the stack. A previous attempt on the block side
didn't pan out and got reverted, hence the need to check for this
information out-of-band right now.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

Merge branches 'acpi-tables' and 'acpi-scan'

* acpi-tables:
  ACPI: tables: x86: Reserve memory occupied by ACPI tables

* acpi-scan:
  ACPI: scan: Fix _STA getting called on devices with unmet dependencies

Merge branch 'pm-cpufreq'

* pm-cpufreq:
  cpufreq: Fix scaling_{available,boost}_frequencies_show() comments

block: don't ignore REQ_NOWAIT for direct IO

If IOCB_NOWAIT is set on submission, then that needs to get propagated to
REQ_NOWAIT on the block side. Otherwise we completely lose this
information, and any issuer of IOCB_NOWAIT IO will potentially end up
blocking on eg request allocation on the storage side.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

regulator: hi6421v600-regulator: move it from staging

This driver is ready for mainstream. Move it out of staging.

Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Acked-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/815b79a4e93f133478d9a5b2dd429526dcfe1dde.1616695231.git.mchehab+huawei@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: hisilicon,hisi-spmi-controller.yaml cleanup schema

Remove some properties already defined at SPMI bus, and
place the type for the spmi-channel.

Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/4e4a80d3a1b84a354d3c40be374a1689ae6c0fb1.1616695231.git.mchehab+huawei@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: remove unused macro RT_PRINT_DATA

remove unused macro declaration: RT_PRINT_DATA

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/a7d11ff4f0c069fac8b66e9c50a846d004ee2a4a.1617267827.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: use print_hex_dump_debug instead of private RT_PRINT_DATA

replace private macro RT_PRINT_DATA with in-kernel helper
print_hex_dump_debug

Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/2ef2c25c928b15f1c4357146efe193afab606d33.1617267827.git.fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: core: remove empty comment

Remove empty comment because it provides no information.

Signed-off-by: Deborah Brouwer <deborahbrouwer3563@gmail.com>
Link: https://lore.kernel.org/r/2362c76457b94895f0b62b59889e80349bb69d8f.1617229359.git.deborahbrouwer3563@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: core: add * to block comments

Add * at the beginning of each line in block comments to conform to the
Linux kernel coding style. Issue detected using checkpatch.

Signed-off-by: Deborah Brouwer <deborahbrouwer3563@gmail.com>
Link: https://lore.kernel.org/r/bd143fff718849274b065a1c71f1aa7f6fc6f40f.1617229359.git.deborahbrouwer3563@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: core: add comma within a comment

Add a comma to separate repeated words in a comment. The comma preserves
the meaning of the comment while also stopping the checkpatch warning:
WARNING: Possible repeated word: 'very'.

Signed-off-by: Deborah Brouwer <deborahbrouwer3563@gmail.com>
Link: https://lore.kernel.org/r/2944d1a0e8769edb489bb336423625a61d314d05.1617229359.git.deborahbrouwer3563@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8188eu: (trivial) remove a duplicate debug print

Keep the one that shows the wakeup capability.

Signed-off-by: Martin Kaiser <martin@kaiser.cx>
Link: https://lore.kernel.org/r/20210329093215.16186-1-martin@kaiser.cx
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8188eu: remove unused function parameter

rtw_usb_if1_init does not use its struct usb_device_id parameter.

Signed-off-by: Martin Kaiser <martin@kaiser.cx>
Link: https://lore.kernel.org/r/20210328175446.28063-1-martin@kaiser.cx
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: camera: Switch to memdup_user_nul()

Use memdup_user_nul() helper instead of open-coding to
simplify the code.

Reported-by: Hulk Robot <hulkci@huawei.com>
Reviewed-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20210401103645.1558813-1-yangyingliang@huawei.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: greybus: arche-platform: Ending line with argument

Remove checkpatch check "CHECK: Lines should not end with a '('"
with argument present in next line.

Signed-off-by: Beatriz Martins de Carvalho <martinsdecarvalhobeatriz@gmail.com>
Link: https://lore.kernel.org/r/20210401090754.116522-1-martinsdecarvalhobeatriz@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

riscv: Make NUMA depend on MMU

NUMA is useless when NOMMU, and it leads some build error,
make it depend on MMU.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>

riscv: remove unneeded semicolon

Eliminate the following coccicheck warning:
./arch/riscv/mm/kasan_init.c:219:2-3: Unneeded semicolon

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>

riscv,entry: fix misaligned base for excp_vect_table

In RV64, the size of each entry in excp_vect_table is 8 bytes. If the
base of the table is not 8-byte aligned, loading an entry in the table
will raise a misaligned exception. Although such exception will be
handled by opensbi/bbl, this still causes performance degradation.

Signed-off-by: Zihao Yu <yuzihao@ict.ac.cn>
Reviewed-by: Anup Patel <anup@brainfault.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>

riscv: evaluate put_user() arg before enabling user access

The <asm/uaccess.h> header has a problem with put_user(a, ptr) if
the 'a' is not a simple variable, such as a function. This can lead
to the compiler producing code as so:

1: enable_user_access()
2: evaluate 'a' into register 'r'
3: put 'r' to 'ptr'
4: disable_user_acess()

The issue is that 'a' is now being evaluated with the user memory
protections disabled. So we try and force the evaulation by assigning
'x' to __val at the start, and hoping the compiler barriers in
 enable_user_access() do the job of ordering step 2 before step 1.

This has shown up in a bug where 'a' sleeps and thus schedules out
and loses the SR_SUM flag. This isn't sufficient to fully fix, but
should reduce the window of opportunity. The first instance of this
we found is in scheudle_tail() where the code does:

$ less -N kernel/sched/core.c

4263  if (current->set_child_tid)
4264         put_user(task_pid_vnr(current), current->set_child_tid);

Here, the task_pid_vnr(current) is called within the block that has
enabled the user memory access. This can be made worse with KASAN
which makes task_pid_vnr() a rather large call with plenty of
opportunity to sleep.

Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
Reported-by: syzbot+e74b94fe601ab9552d69@syzkaller.appspotmail.com
Suggested-by: Arnd Bergman <arnd@arndb.de>
--
Changes since v1:
- fixed formatting and updated the patch description with more info

Changes since v2:
- fixed commenting on __put_user() (schwab@linux-m68k.org)

Change since v3:
- fixed RFC in patch title. Should be ready to merge.

Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>

riscv: Drop const annotation for sp

The const annotation should not be used for 'sp', or it will
become read only and lead to bad stack output.

Fixes: dec822771b01 ("riscv: stacktrace: Move register keyword to beginning of declaration")
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>

Merge tag 'lto-v5.12-rc6' of git://git./linux/kernel/git/kees/linux

Pull LTO fix from Kees Cook:
 "It seems that there is a bug in ld.bfd when doing module section
  merging.

  As explicit merging is only needed for LTO, the work-around is to only
  do it under LTO, leaving the original section layout choices alone
  under normal builds:

   - Only perform explicit module section merges under LTO (Sean
     Christopherson)"

* tag 'lto-v5.12-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
  kbuild: lto: Merge module sections if and only if CONFIG_LTO_CLANG is enabled

kbuild: lto: Merge module sections if and only if CONFIG_LTO_CLANG is enabled

Merge module sections only when using Clang LTO. With ld.bfd, merging
sections does not appear to update the symbol tables for the module,
e.g. 'readelf -s' shows the value that a symbol would have had, if
sections were not merged. ld.lld does not show this problem.

The stale symbol table breaks gdb's function disassembler, and presumably
other things, e.g.

  gdb -batch -ex "file arch/x86/kvm/kvm.ko" -ex "disassemble kvm_init"

reads the wrong bytes and dumps garbage.

Fixes: dd2776222abb ("kbuild: lto: merge module sections")
Cc: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Reviewed-by: Sami Tolvanen <samitolvanen@google.com>
Tested-by: Sami Tolvanen <samitolvanen@google.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20210322234438.502582-1-seanjc@google.com

Merge tag 'for-linus' of git://git./virt/kvm/kvm

Pull kvm fixes from Paolo Bonzini:
 "It's a bit larger than I (and probably you) would like by the time we
  get to -rc6, but perhaps not entirely unexpected since the changes in
  the last merge window were larger than usual.

  x86:
   - Fixes for missing TLB flushes with TDP MMU

   - Fixes for race conditions in nested SVM

   - Fixes for lockdep splat with Xen emulation

   - Fix for kvmclock underflow

   - Fix srcdir != builddir builds

   - Other small cleanups

  ARM:
   - Fix GICv3 MMIO compatibility probing

   - Prevent guests from using the ARMv8.4 self-hosted tracing
     extension"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  selftests: kvm: Check that TSC page value is small after KVM_SET_CLOCK(0)
  KVM: x86: Prevent 'hv_clock->system_time' from going negative in kvm_guest_time_update()
  KVM: x86: disable interrupts while pvclock_gtod_sync_lock is taken
  KVM: x86: reduce pvclock_gtod_sync_lock critical sections
  KVM: SVM: ensure that EFER.SVME is set when running nested guest or on nested vmexit
  KVM: SVM: load control fields from VMCB12 before checking them
  KVM: x86/mmu: Don't allow TDP MMU to yield when recovering NX pages
  KVM: x86/mmu: Ensure TLBs are flushed for TDP MMU during NX zapping
  KVM: x86/mmu: Ensure TLBs are flushed when yielding during GFN range zap
  KVM: make: Fix out-of-source module builds
  selftests: kvm: make hardware_disable_test less verbose
  KVM: x86/vPMU: Forbid writing to MSR_F15H_PERF MSRs when guest doesn't have X86_FEATURE_PERFCTR_CORE
  KVM: x86: remove unused declaration of kvm_write_tsc()
  KVM: clean up the unused argument
  tools/kvm_stat: Add restart delay
  KVM: arm64: Fix CPU interface MMIO compatibility detection
  KVM: arm64: Disable guest access to trace filter controls
  KVM: arm64: Hide system instruction access to Trace registers

Merge tag 'drm-fixes-2021-04-02' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "Things have settled down in time for Easter, a random smattering of
  small fixes across a few drivers.

  I'm guessing though there might be some i915 and misc fixes out there
  I haven't gotten yet, but since today is a public holiday here, I'm
  sending this early so I can have the day off, I'll see if more
  requests come in and decide what to do with them later.

  amdgpu:
   - Polaris idle power fix
   - VM fix
   - Vangogh S3 fix
   - Fixes for non-4K page sizes

  amdkfd:
   - dqm fence memory corruption fix

  tegra:
   - lockdep warning fix
   - runtine PM reference fix
   - display controller fix
   - PLL Fix

  imx:
   - memory leak in error path fix
   - LDB driver channel registration fix
   - oob array warning in LDB driver

  exynos
   - unused header file removal"

* tag 'drm-fixes-2021-04-02' of git://anongit.freedesktop.org/drm/drm:
  drm/amdgpu: check alignment on CPU page for bo map
  drm/amdgpu: Set a suitable dev_info.gart_page_size
  drm/amdgpu/vangogh: don't check for dpm in is_dpm_running when in suspend
  drm/amdkfd: dqm fence memory corruption
  drm/tegra: sor: Grab runtime PM reference across reset
  drm/tegra: dc: Restore coupling of display controllers
  gpu: host1x: Use different lock classes for each client
  drm/tegra: dc: Don't set PLL clock to 0Hz
  drm/amdgpu: fix offset calculation in amdgpu_vm_bo_clear_mappings()
  drm/amd/pm: no need to force MCLK to highest when no display connected
  drm/exynos/decon5433: Remove the unused include statements
  drm/imx: imx-ldb: fix out of bounds array access warning
  drm/imx: imx-ldb: Register LDB channel1 when it is the only channel to be used
  drm/imx: fix memory leak when fails to init

Merge tag 'imx-drm-fixes-2021-04-01' of git://git.pengutronix.de/git/pza/linux into drm-fixes

drm/imx: imx-drm-core and imx-ldb fixes

Fix a memory leak in an error path during DRM device initialization,
fix the LDB driver to register channel 1 even if channel 0 is unused,
and fix an out of bounds array access warning in the LDB driver.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Philipp Zabel <p.zabel@pengutronix.de>
Link: https://patchwork.freedesktop.org/patch/msgid/20210401092235.GA13586@pengutronix.de

Merge tag 'drm/tegra/for-5.12-rc6' of ssh://git.freedesktop.org/git/tegra/linux into drm-fixes

drm/tegra: Fixes for v5.12-rc6

This contains a couple of fixes for various issues such as lockdep
warnings, runtime PM references, coupled display controllers and
misconfigured PLLs.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thierry Reding <thierry.reding@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210401163352.3348296-1-thierry.reding@gmail.com

tracing: Fix stack trace event size

Commit cbc3b92ce037 fixed an issue to modify the macros of the stack trace
event so that user space could parse it properly. Originally the stack
trace format to user space showed that the called stack was a dynamic
array. But it is not actually a dynamic array, in the way that other
dynamic event arrays worked, and this broke user space parsing for it. The
update was to make the array look to have 8 entries in it. Helper
functions were added to make it parse it correctly, as the stack was
dynamic, but was determined by the size of the event stored.

Although this fixed user space on how it read the event, it changed the
internal structure used for the stack trace event. It changed the array
size from [0] to [8] (added 8 entries). This increased the size of the
stack trace event by 8 words. The size reserved on the ring buffer was the
size of the stack trace event plus the number of stack entries found in
the stack trace. That commit caused the amount to be 8 more than what was
needed because it did not expect the caller field to have any size. This
produced 8 entries of garbage (and reading random data) from the stack
trace event:

          <idle>-0       [002] d... 1976396.837549: <stack trace>
 => trace_event_raw_event_sched_switch
 => __traceiter_sched_switch
 => __schedule
 => schedule_idle
 => do_idle
 => cpu_startup_entry
 => secondary_startup_64_no_verify
 => 0xc8c5e150ffff93de
 => 0xffff93de
 => 0
 => 0
 => 0xc8c5e17800000000
 => 0x1f30affff93de
 => 0x00000004
 => 0x200000000

Instead, subtract the size of the caller field from the size of the event
to make sure that only the amount needed to store the stack trace is
reserved.

Link: https://lore.kernel.org/lkml/your-ad-here.call-01617191565-ext-9692@work.hours/
Cc: stable@vger.kernel.org
Fixes: cbc3b92ce037 ("tracing: Set kernel_stack's caller size properly")
Reported-by: Vasily Gorbik <gor@linux.ibm.com>
Tested-by: Vasily Gorbik <gor@linux.ibm.com>
Acked-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>