staging: most-core: Fix checkpatch warnings

Hi,

This patch solves the following checkpatch.pl's messages in drivers/staging/most/core.c.

WARNING: line over 80 characters
+ return snprintf(buf, PAGE_SIZE, "%s", ch_data_type[i].name);

CHECK: Please use a blank line after function/struct/union/enum declarations
+}
+/**

Signed-off-by: Peikan Tsai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192u: Add space around binary operators

This patch fixes the file r8180_93cx6.c to avoid the checkpatch.pl warnings:

CHECK: spaces preferred around that '<<' (ctx:VxV)
CHECK: spaces preferred around that '-' (ctx:VxV)

Signed-off-by: Sumera Priyadarsini <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192u: remove code under TO_DO_LIST

Several blocks of code are guarded by #ifdef TO_DO_LIST. If this is
defined, compilation fails. No machinery exists to define this, and no
documenation on the in-progress feature exists. Since this code is dead,
let's delete it.

Signed-off-by: Stephen Brennan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192u: Remove unnecessary blank lines

This patch fixes the file r8180_93cx6.c to avoid the checkpatch.pl warning:

CHECK: Please don't use multiple blank lines

Signed-off-by: Sumera Priyadarsini <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192u: remove redundant assignment to pointer crypt

The pointer crypt is being set with a value that is never read,
the assignment is redundant and hence can be removed.

Thanks to Dan Carpenter for sanity checking that this was indeed
redundant.

Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <[email protected]>
Reviewed-by: Dan Carpenter <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8723bs: os_dep: Drop condition with no effect

As the "else if" and "else" branch body are identical the condition
has no effect. So drop the "else if" condition.

Signed-off-by: Saurav Girepunje <[email protected]>
Link: https://lore.kernel.org/r/20190821181631.GA11082@saurav
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8723bs: hal: Drop condition with no effect

As the "else if" and "else" branch body are identical the condition
has no effect. So drop the else if condition.

Signed-off-by: Saurav Girepunje <[email protected]>
Link: https://lore.kernel.org/r/20190821180153.GA10678@saurav
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8723bs: os_dep: Remove unused declarations

Remove unused Macro declarations

Signed-off-by: Hariprasad Kelam <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

erofs: move erofs out of staging

EROFS filesystem has been merged into linux-staging for a year.

EROFS is designed to be a better solution of saving extra storage
space with guaranteed end-to-end performance for read-only files
with the help of reduced metadata, fixed-sized output compression
and decompression inplace technologies.

In the past year, EROFS was greatly improved by many people as
a staging driver, self-tested, betaed by a large number of our
internal users, successfully applied to almost all in-service
HUAWEI smartphones as the part of EMUI 9.1 and proven to be stable
enough to be moved out of staging.

EROFS is a self-contained filesystem driver. Although there are
still some TODOs to be more generic, we have a dedicated team
actively keeping on working on EROFS in order to make it better
with the evolution of Linux kernel as the other in-kernel filesystems.

As Pavel suggested, it's better to do as one commit since git
can do moves and all histories will be saved in this way.

Let's promote it from staging and enhance it more actively as
a "real" part of kernel for more wider scenarios!

Cc: Greg Kroah-Hartman <[email protected]>
Cc: Alexander Viro <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Stephen Rothwell <[email protected]>
Cc: Theodore Ts'o <[email protected]>
Cc: Pavel Machek <[email protected]>
Cc: David Sterba <[email protected]>
Cc: Amir Goldstein <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Darrick J . Wong <[email protected]>
Cc: Dave Chinner <[email protected]>
Cc: Jaegeuk Kim <[email protected]>
Cc: Jan Kara <[email protected]>
Cc: Richard Weinberger <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Chao Yu <[email protected]>
Cc: Miao Xie <[email protected]>
Cc: Li Guifu <[email protected]>
Cc: Fang Wei <[email protected]>
Signed-off-by: Gao Xiang <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192u: fix spacing in ieee80211

Checkpatch emits several errors, warnings, and checks about spacing.
Apply checkpatch's suggested spacing rules.

Signed-off-by: Stephen Brennan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192u: fix macro alignment in ieee80211

Several macros display unaligned, due to mixes of tabs and spaces. These
can be fixed by making spacing consistent, do this.

Signed-off-by: Stephen Brennan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192u: fix OPEN_BRACE errors in ieee80211

Checkpatch emits several errors regarding braces being on the incorrect
line. These can be fixed by moving the brace, do this. In a few cases,
some comments were moved to facilitate this.

Signed-off-by: Stephen Brennan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192e: remove set but not used variable 'data_len'

Fixes gcc '-Wunused-but-set-variable' warning:

 In function ieee80211_ccmp_encrypt:
drivers/staging/rtl8192u/ieee80211/ieee80211_crypt_ccmp.c:162:6:
 warning: variable data_len set but not used [-Wunused-but-set-variable]

It is not used since commit 5ee5265674ce ("staging:
rtl8192e: rtllib_crypt_ccmp.c: Use crypto API ccm(aes)")

Reported-by: Hulk Robot <[email protected]>
Signed-off-by: YueHaibing <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192e: remove two set but not used variables

Fixes gcc '-Wunused-but-set-variable' warning:

 In function '_rtl92e_dm_tx_power_tracking_callback_tssi':
drivers/staging/rtl8192e/rtl8192e/rtl_dm.c:621:7:
 warning: variable 'bHighpowerstate' set but not used [-Wunused-but-set-variable]
 In function '_rtl92e_dm_rx_path_sel_byrssi':
drivers/staging/rtl8192e/rtl8192e/rtl_dm.c:1904:32:
 warning: variable 'cck_rx_ver2_min_index' set but not used [-Wunused-but-set-variable]

They are never used, so can be removed.

Reported-by: Hulk Robot <[email protected]>
Signed-off-by: YueHaibing <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192u: ieee80211: remove set but not used variable 'data_len'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/staging/rtl8192u/ieee80211/ieee80211_crypt_ccmp.c:
 In function ieee80211_ccmp_encrypt:
drivers/staging/rtl8192u/ieee80211/ieee80211_crypt_ccmp.c:162:6:
 warning: variable data_len set but not used [-Wunused-but-set-variable]

It is not used since commit eb0e7bf3ca94 ("staging:
rtl8192u: ieee80211: ieee80211_crypt_ccmp.c: Use crypto API ccm(aes)")

Reported-by: Hulk Robot <[email protected]>
Signed-off-by: YueHaibing <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: detect potential multiref due to corrupted images

As reported by erofs-utils fuzzer, currently, multiref
(ondisk deduplication) hasn't been supported for now,
we should forbid it properly.

Fixes: 3883a79abd02 ("staging: erofs: introduce VLE decompression support")
Cc: <[email protected]> # 4.19+
Signed-off-by: Gao Xiang <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rts5208: remove redundant assignment to retval

Variable retval is initialized to a value that is never read and it
is re-assigned later. The initialization is redundant and can be
removed.

Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: kpc2000: kpc2000_i2c: Fix different address spaces warnings

This patch fixes the following sparse warnings:

kpc2000_i2c.c:137: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:137:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:137:    got void *
kpc2000_i2c.c:146: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:146:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:146:    got void *
kpc2000_i2c.c:147: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:147:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:147:    got void *
kpc2000_i2c.c:166: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:166:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:166:    got void *
kpc2000_i2c.c:166: warning: incorrect type in argument 2
                                  (different address spaces)
kpc2000_i2c.c:166:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:166:    got void *
kpc2000_i2c.c:168: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:168:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:168:    got void *
kpc2000_i2c.c:168: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:168:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:168:    got void *
kpc2000_i2c.c:171: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:171:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:171:    got void *
kpc2000_i2c.c:174: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:174:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:174:    got void *
kpc2000_i2c.c:193: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:193:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:193:    got void *
kpc2000_i2c.c:194: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:194:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:194:    got void *
kpc2000_i2c.c:214: warning: incorrect type in argument 2
                                  (different address spaces)
kpc2000_i2c.c:214:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:214:    got void *
kpc2000_i2c.c:219: warning: incorrect type in argument 1
                                  (different address spaces)
kpc2000_i2c.c:219:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:219:    got void *
kpc2000_i2c.c:226: warning: incorrect type in argument 2
                                  (different address spaces)
kpc2000_i2c.c:226:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:226:    got void *
kpc2000_i2c.c:238: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:238:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:238:    got void *
kpc2000_i2c.c:244: warning: incorrect type in argument 2
                                  (different address spaces)
kpc2000_i2c.c:244:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:244:    got void *
kpc2000_i2c.c:252: warning: incorrect type in argument 1
                                  (different address spaces)
kpc2000_i2c.c:252:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:252:    got void *
kpc2000_i2c.c:257: warning: incorrect type in argument 2
                                  (different address spaces)
kpc2000_i2c.c:257:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:257:    got void *
kpc2000_i2c.c:259: warning: incorrect type in argument 2
   (different address spaces)
kpc2000_i2c.c:259:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:259:    got void *
kpc2000_i2c.c:267: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:267:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:267:    got void *
kpc2000_i2c.c:273: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:273:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:273:    got void *
kpc2000_i2c.c:293: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:293:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:293:    got void *
kpc2000_i2c.c:294: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:294:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:294:    got void *
kpc2000_i2c.c:309: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:309:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:309:    got void *
kpc2000_i2c.c:312: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:312:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:312:    got void *
kpc2000_i2c.c:317: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:317:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:317:    got void *
kpc2000_i2c.c:324: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:324:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:324:    got void *
kpc2000_i2c.c:328: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:328:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:328:    got void *
kpc2000_i2c.c:329: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:329:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:329:    got void *
kpc2000_i2c.c:330: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:330:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:330:    got void *
kpc2000_i2c.c:338: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:338:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:338:    got void *
kpc2000_i2c.c:340: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:340:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:340:    got void *
kpc2000_i2c.c:342: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:342:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:342:    got void *
kpc2000_i2c.c:350: warning: incorrect type in argument 1
                                  (different address spaces)
kpc2000_i2c.c:350:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:350:    got void *
kpc2000_i2c.c:350: warning: incorrect type in argument 2
                                  (different address spaces)
kpc2000_i2c.c:350:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:350:    got void *
kpc2000_i2c.c:351: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:351:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:351:    got void *
kpc2000_i2c.c:414: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:414:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:414:    got void *
kpc2000_i2c.c:420: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:420:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:420:    got void *
kpc2000_i2c.c:422: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:422:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:422:    got void *
kpc2000_i2c.c:427: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:427:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:427:    got void *
kpc2000_i2c.c:428: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:428:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:428:    got void *
kpc2000_i2c.c:430: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:430:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:430:    got void *
kpc2000_i2c.c:435: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:435:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:435:    got void *
kpc2000_i2c.c:436: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:436:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:436:    got void *
kpc2000_i2c.c:438: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:438:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:438:    got void *
kpc2000_i2c.c:439: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:439:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:439:    got void *
kpc2000_i2c.c:445: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:445:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:445:    got void *
kpc2000_i2c.c:446: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:446:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:446:    got void *
kpc2000_i2c.c:454: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:454:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:454:    got void *
kpc2000_i2c.c:459: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:459:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:459:    got void *
kpc2000_i2c.c:461: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:461:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:461:    got void *
kpc2000_i2c.c:472: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:472:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:472:    got void *
kpc2000_i2c.c:472: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:472:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:472:    got void *
kpc2000_i2c.c:475: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:475:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:475:    got void *
kpc2000_i2c.c:475: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:475:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:475:    got void *
kpc2000_i2c.c:493: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:493:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:493:    got void *
kpc2000_i2c.c:493: warning: incorrect type in argument 2
                                   (different address spaces)
kpc2000_i2c.c:493:    expected void volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:493:    got void *
kpc2000_i2c.c:512: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:512:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:512:    got void *
kpc2000_i2c.c:516: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:516:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:516:    got void *
kpc2000_i2c.c:516: warning: incorrect type in argument 1
                                   (different address spaces)
kpc2000_i2c.c:516:    expected void const volatile [noderef] <asn:2> *addr
kpc2000_i2c.c:516:    got void *

Signed-off-by: Eduardo Barretto <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: android: Remove ion device tree bindings from the TODO

Commit 23a4388f24f5 ("staging: android: ion: Remove file ion_chunk_heap.c")
and eadbf7a34e44 ("staging: android: ion: Remove file ion_carveout_heap.c")
removed the chunk and carveout heaps from ion but left behind the device
tree bindings for them in the TODO, this patch removes it.

Signed-off-by: Donald Yandt <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8712: Improve naming of include hearder guards

Choose a better name for the include hearder guard used in rtl871x_io.h.
'_IO_H_' is to generic and does not match the comment after the #endif.
Use '_RTL871X_IO_H_' instead.

Also make the comments in the #endif /* XXX */ match the name used in
#ifndef.

Signed-off-by: Christophe JAILLET <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: avoid loop in submit chains

As reported by erofs-utils fuzzer, 2 conditions
can happen in corrupted images, which can cause
unexpected behaviors.
 - access the same pcluster one more time;
 - access the tail end pcluster again, e.g.
            _ access again (will trigger tail merging)
           |
     1 2 3 1 2             ->   1 2 3 1
     |_ tail end of the chain    \___/ (unexpected behavior)
Let's detect and avoid them now.

Reviewed-by: Chao Yu <[email protected]>
Signed-off-by: Gao Xiang <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8723bs: core: Remove unneeded declaration WFD_OUI

Remove unneeded declaration "extern unsigned char WFD_OUI"

Signed-off-by: Hariprasad Kelam <[email protected]>
Link: https://lore.kernel.org/r/20190819165705.GA5782@hari-Inspiron-1545
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: comedi: ni_mio_common: Fix a typo in ni_mio_common.c

This patch fix a spelling typo in ni_mio_common.c

Signed-off-by: Masanari Iida <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: avoid endless loop of invalid lookback distance 0

As reported by erofs-utils fuzzer, Lookback distance should
be a positive number, so it should be actually looked back
rather than spinning.

Fixes: 02827e1796b3 ("staging: erofs: add erofs_map_blocks_iter")
Cc: <[email protected]> # 4.19+
Signed-off-by: Gao Xiang <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: add two missing erofs_workgroup_put for corrupted images

As reported by erofs-utils fuzzer, these error handling
path will be entered to handle corrupted images.

Lack of erofs_workgroup_puts will cause unmounting
unsuccessfully.

Fix these return values to EFSCORRUPTED as well.

Fixes: 3883a79abd02 ("staging: erofs: introduce VLE decompression support")
Cc: <[email protected]> # 4.19+
Signed-off-by: Gao Xiang <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: cannot set EROFS_V_Z_INITED_BIT if fill_inode_lazy fails

As reported by erofs-utils fuzzer, unsupported compressed
clustersize will make fill_inode_lazy fail, for such case
we cannot set EROFS_V_Z_INITED_BIT since we need return
failure for each z_erofs_map_blocks_iter().

Fixes: 152a333a5895 ("staging: erofs: add compacted compression indexes support")
Cc: <[email protected]> # 5.3+
Signed-off-by: Gao Xiang <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: some compressed cluster should be submitted for corrupted images

As reported by erofs_utils fuzzer, a logical page can belong
to at most 2 compressed clusters, if one compressed cluster
is corrupted, but the other has been ready in submitting chain.

The chain needs to submit anyway in order to keep the page
working properly (page unlocked with PG_error set, PG_uptodate
not set).

Let's fix it now.

Fixes: 3883a79abd02 ("staging: erofs: introduce VLE decompression support")
Cc: <[email protected]> # 4.19+
Signed-off-by: Gao Xiang <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: fix an error handling in erofs_readdir()

Richard observed a forever loop of erofs_read_raw_page() [1]
which can be generated by forcely setting ->u.i_blkaddr
to 0xdeadbeef (as my understanding block layer can
handle access beyond end of device correctly).

After digging into that, it seems the problem is highly
related with directories and then I found the root cause
is an improper error handling in erofs_readdir().

Let's fix it now.

[1] https://lore.kernel.org/r/1163995781.68824.1566084358245[email protected]/

Reported-by: Richard Weinberger <[email protected]>
Fixes: 3aa8ec716e52 ("staging: erofs: add directory operations")
Cc: <[email protected]> # 4.19+
Reviewed-by: Chao Yu <[email protected]>
Signed-off-by: Gao Xiang <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: refuse to mount images with malformed volume name

As Richard reminder [1], A valid volume name should be
ended in NIL terminator within the length of volume_name.

Since this field currently isn't really used, let's fix
it to avoid potential bugs in the future.

[1] https://lore.kernel.org/r/1133002215.69049.1566119033047[email protected]/

Reported-by: Richard Weinberger <[email protected]>
Signed-off-by: Gao Xiang <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Merge 5.3-rc5 into staging-next

We need the staging fixes in here as well.

Signed-off-by: Greg Kroah-Hartman <[email protected]>

Linux 5.3-rc5

Merge tag 'fixes-for-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux

Pull MTD fix from Richard Weinberger:
 "A single fix for MTD to correctly set the spi-nor WP pin"

* tag 'fixes-for-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux:
  mtd: spi-nor: Fix the disabling of write protection at init

Merge tag 'for-5.3-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:
 "Two fixes that popped up during testing:

   - fix for sysfs-related code that adds/removes block groups, warnings
     appear during several fstests in connection with sysfs updates in
     5.3, the fix essentially replaces a workaround with scope NOFS and
     applies to 5.2-based branch too

   - add sanity check of trim range"

* tag 'for-5.3-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: trim: Check the range passed into to prevent overflow
  Btrfs: fix sysfs warning and missing raid sysfs directories

Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 fixes from Thomas Gleixner:
 "A set of fixes for x86:

   - Fix the inconsistent error handling in the umwait init code

   - Rework the boot param zeroing so gcc9 stops complaining about out
     of bound memset. The resulting source code is actually more sane to
     read than the smart solution we had

   - Maintainers update so Tony gets involved when Intel models are
     added

   - Some more fallthrough fixes"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/boot: Save fields explicitly, zero out everything else
  MAINTAINERS, x86/CPU: Tony Luck will maintain asm/intel-family.h
  x86/fpu/math-emu: Address fallthrough warnings
  x86/apic/32: Fix yet another implicit fallthrough warning
  x86/umwait: Fix error handling in umwait_init()

Merge branch 'efi-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull EFI fix from Thomas Gleixner:
 "A single fix for a EFI mixed mode regression caused by recent rework
  which did not take the firmware bitwidth into account"

* 'efi-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  efi-stub: Fix get_efi_config_table on mixed-mode setups

Merge tag 'spdx-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/spdx

Pull SPDX fixes from Greg KH:
 "Here are four small SPDX fixes for 5.3-rc5.

  A few style fixes for some SPDX comments, added an SPDX tag for one
  file, and fix up some GPL boilerplate for another file.

  All of these have been in linux-next for a few weeks with no reported
  issues (they are comment changes only, so that's to be expected...)"

* tag 'spdx-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/spdx:
  i2c: stm32: Use the correct style for SPDX License Identifier
  intel_th: Use the correct style for SPDX License Identifier
  coccinelle: api/atomic_as_refcounter: add SPDX License Identifier
  kernel/configs: Replace GPL boilerplate code with SPDX identifier

Merge tag 'char-misc-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc

Pull char/misc driver fixes from Greg KH:
 "Here are some small char and misc driver fixes for 5.3-rc5.

  These are two different subsystems needing some fixes, the habanalabs
  driver which is has some more big endian fixes for problems found. The
  other are some small soundwire fixes, including some Kconfig
  dependencies needed to resolve reported build errors.

  All of these have been in linux-next this week with no reported
  issues"

* tag 'char-misc-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  misc: xilinx-sdfec: fix dependency and build error
  habanalabs: fix device IRQ unmasking for BE host
  habanalabs: fix endianness handling for internal QMAN submission
  habanalabs: fix completion queue handling when host is BE
  habanalabs: fix endianness handling for packets from user
  habanalabs: fix DRAM usage accounting on context tear down
  habanalabs: Avoid double free in error flow
  soundwire: fix regmap dependencies and align with other serial links
  soundwire: cadence_master: fix definitions for INTSTAT0/1
  soundwire: cadence_master: fix register definition for SLAVE_STATE

Merge tag 'staging-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging

Pull staging/IIO fixes from Greg KH:
 "Here are four small staging and iio driver fixes for 5.3-rc5

  Two are for the dt3000 comedi driver for some reported problems found
  in that codebase, and two are some small iio fixes.

  All of these have been in linux-next this week with no reported
  issues"

* tag 'staging-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  staging: comedi: dt3000: Fix rounding up of timer divisor
  staging: comedi: dt3000: Fix signed integer overflow 'divider * base'
  iio: adc: max9611: Fix temperature reading in probe
  iio: frequency: adf4371: Fix output frequency setting

Merge tag 'usb-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb

Pull USB fixes from Greg KH:
 "Here are number of small USB fixes for 5.3-rc5.

  Syzbot has been on a tear recently now that it has some good USB
  debugging hooks integrated, so there's a number of fixes in here found
  by those tools for some _very_ old bugs. Also a handful of gadget
  driver fixes for reported issues, some hopefully-final dma fixes for
  host controller drivers, and some new USB serial gadget driver ids.

  All of these have been in linux-next this week with no reported issues
  (the usb-serial ones were in linux-next in its own branch, but merged
  into mine on Friday)"

* tag 'usb-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
  usb: add a hcd_uses_dma helper
  usb: don't create dma pools for HCDs with a localmem_pool
  usb: chipidea: imx: fix EPROBE_DEFER support during driver probe
  usb: host: fotg2: restart hcd after port reset
  USB: CDC: fix sanity checks in CDC union parser
  usb: cdc-acm: make sure a refcount is taken early enough
  USB: serial: option: add the BroadMobi BM818 card
  USB: serial: option: Add Motorola modem UARTs
  USB: core: Fix races in character device registration and deregistraion
  usb: gadget: mass_storage: Fix races between fsg_disable and fsg_set_alt
  usb: gadget: composite: Clear "suspended" on reset/disconnect
  usb: gadget: udc: renesas_usb3: Fix sysfs interface of "role"
  USB: serial: option: add D-Link DWM-222 device ID
  USB: serial: option: Add support for ZTE MF871A

staging: gasket: Remove unnecessary line-breaks in function signatures

This patch fixes the function signatures for gasket_read_page_table_size,
gasket_read_simple_page_table_size, gasket_partition_page_table,
gasket_config_coherent_allocator to avoid  the checkpatch.pl warning:

CHECK: Lines should not end with a '('

Signed-off-by: Sumera Priyadarsini <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: use common file type conversion

Deduplicate the EROFS file type conversion implementation and
remove EROFS_FT_* definitions since it's the same as defined
by POSIX, let's follow ext2 as Linus pointed out [1]
commit e10892189428 ("ext2: use common file type conversion").

[1] https://lore.kernel.org/r/CAHk-=wiUs+b=iVKM3mVooXgVk7cmmC67KTmnAuL0cd_cMMVAKw@mail.gmail.com/

Reported-by: Linus Torvalds <[email protected]>
Signed-off-by: Gao Xiang <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192e: rtllib_crypt_ccmp.c: Use crypto API ccm(aes)

Use ccm(aes) aead transform instead of invoking the AES block cipher
block by block.

Signed-off-by: Christina Quast <[email protected]>
Reviewed-by: Ard Biesheuvel <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192u: ieee80211: ieee80211_crypt_ccmp.c: Use crypto API ccm(aes)

Use ccm(aes) aead transform instead of invoking the AES block cipher
block by block.

Signed-off-by: Christina Quast <[email protected]>
Reviewed-by: Ard Biesheuvel <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Merge tag 'for-linus-2019-08-17' of git://git.kernel.dk/linux-block

Pull block fixes from Jens Axboe:
 "A collection of fixes that should go into this series. This contains:

   - Revert of the REQ_NOWAIT_INLINE and associated dio changes. There
     were still corner cases there, and even though I had a solution for
     it, it's too involved for this stage. (me)

   - Set of NVMe fixes (via Sagi)

   - io_uring fix for fixed buffers (Anthony)

   - io_uring defer issue fix (Jackie)

   - Regression fix for queue sync at exit time (zhengbin)

   - xen blk-back memory leak fix (Wenwen)"

* tag 'for-linus-2019-08-17' of git://git.kernel.dk/linux-block:
  io_uring: fix an issue when IOSQE_IO_LINK is inserted into defer list
  block: remove REQ_NOWAIT_INLINE
  io_uring: fix manual setup of iov_iter for fixed buffers
  xen/blkback: fix memory leaks
  blk-mq: move cancel of requeue_work to the front of blk_exit_queue
  nvme-pci: Fix async probe remove race
  nvme: fix controller removal race with scan work
  nvme-rdma: fix possible use-after-free in connect error flow
  nvme: fix a possible deadlock when passthru commands sent to a multipath device
  nvme-core: Fix extra device_put() call on error path
  nvmet-file: fix nvmet_file_flush() always returning an error
  nvmet-loop: Flush nvme_delete_wq when removing the port
  nvmet: Fix use-after-free bug when a port is removed
  nvme-multipath: revalidate nvme_ns_head gendisk in nvme_validate_ns

Merge tag 'hyperv-fixes-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux

Pull Hyper-V fixes from Sasha Levin:

 - A few fixes for the userspace hyper-v tools from Adrian Vladu.

 - A fix for the hyper-v MAINTAINERs entry from Lan Tianyu.

 - Fix for SPDX license identifier in the userspace tools from Nishad
   Kamdar.

* tag 'hyperv-fixes-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/hyperv/linux:
  MAINTAINERS: Fix Hyperv vIOMMU driver file name
  tools: hv: Use the correct style for SPDX License Identifier
  tools: hv: fix typos in toolchain
  tools: hv: fix KVP and VSS daemons exit code
  tools: hv: fixed Python pep8/flake8 warnings for lsvmbus

MAINTAINERS: Fix Hyperv vIOMMU driver file name

The Hyperv vIOMMU file name should be "hyperv-iommu.c" rather
than "hyperv_iommu.c". This patch is to fix it.

Signed-off-by: Lan Tianyu <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>

tools: hv: Use the correct style for SPDX License Identifier

This patch corrects the SPDX License Identifier style
in the trace header file related to Microsoft Hyper-V
client drivers.
For C header files Documentation/process/license-rules.rst
mandates C-like comments (opposed to C source files where
C++ style should be used)

Changes made by using a script provided by Joe Perches here:
https://lkml.org/lkml/2019/2/7/46

Suggested-by: Joe Perches <[email protected]>
Signed-off-by: Nishad Kamdar <[email protected]>
Acked-by: Greg Kroah-Hartman <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>

tools: hv: fix typos in toolchain

Fix typos in the HyperV toolchain.

Signed-off-by: Adrian Vladu <[email protected]>
Cc: "K. Y. Srinivasan" <[email protected]>
Cc: Haiyang Zhang <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Sasha Levin <[email protected]>
Cc: Alessandro Pilotti <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>

tools: hv: fix KVP and VSS daemons exit code

HyperV KVP and VSS daemons should exit with 0 when the '--help'
or '-h' flags are used.

Signed-off-by: Adrian Vladu <[email protected]>
Cc: "K. Y. Srinivasan" <[email protected]>
Cc: Haiyang Zhang <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Sasha Levin <[email protected]>
Cc: Alessandro Pilotti <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>

tools: hv: fixed Python pep8/flake8 warnings for lsvmbus

Fixed pep8/flake8 python style code for lsvmbus tool.

The TAB indentation was on purpose ignored (pep8 rule W191) to make
sure the code is complying with the Linux code guideline.
The following command doe not show any warnings now:
pep8 --ignore=W191 lsvmbus
flake8 --ignore=W191 lsvmbus

Signed-off-by: Adrian Vladu <[email protected]>
Cc: "K. Y. Srinivasan" <[email protected]>
Cc: Haiyang Zhang <[email protected]>
Cc: Stephen Hemminger <[email protected]>
Cc: Sasha Levin <[email protected]>
Cc: Dexuan Cui <[email protected]>
Cc: Alessandro Pilotti <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>

Merge branch 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux

Pull i2c fixes from Wolfram Sang:
 "I2C has one revert because of a regression, two fixes for tiny race
  windows (which we were not able to trigger), a MAINTAINERS addition,
  and a SPDX fix"

* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  i2c: stm32: Use the correct style for SPDX License Identifier
  i2c: emev2: avoid race when unregistering slave client
  i2c: rcar: avoid race when unregistering slave client
  MAINTAINERS: i2c-imx: take over maintainership
  Revert "i2c: imx: improve the error handling in i2c_imx_dma_request()"

Merge tag 'riscv/for-v5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux

Pull RISC-V fixes from Paul Walmsley:

 - Two patches to fix significant bugs in floating point register
   context handling

 - A minor fix in RISC-V flush_tlb_page(), to supply a valid end address
   to flush_tlb_range()

 - Two minor defconfig additions: to build the virtio hwrng driver by
   default (for QEMU targets), and to partially synchronize the 32-bit
   defconfig with the 64-bit defconfig

* tag 'riscv/for-v5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
  riscv: Make __fstate_clean() work correctly.
  riscv: Correct the initialized flow of FP register
  riscv: defconfig: Update the defconfig
  riscv: rv32_defconfig: Update the defconfig
  riscv: fix flush_tlb_range() end address for flush_tlb_page()

Merge tag 'usb-serial-5.3-rc5' of https://git.kernel.org/pub/scm/linux/kernel/git/johan/usb-serial into usb-linus

Johan writes:

USB-serial fixes for 5.3-rc5

Here are some new modem device ids.

All have been in linux-next with no reported issues.

Signed-off-by: Johan Hovold <[email protected]>
* tag 'usb-serial-5.3-rc5' of https://git.kernel.org/pub/scm/linux/kernel/git/johan/usb-serial:
  USB: serial: option: add the BroadMobi BM818 card
  USB: serial: option: Add Motorola modem UARTs
  USB: serial: option: add D-Link DWM-222 device ID
  USB: serial: option: Add support for ZTE MF871A

Merge tag 'xtensa-20190816' of git://github.com/jcmvbkbc/linux-xtensa

Pull Xtensa fix from Max Filippov:
 "Add missing isync into cpu_reset to make sure ITLB changes are
  effective"

* tag 'xtensa-20190816' of git://github.com/jcmvbkbc/linux-xtensa:
  xtensa: add missing isync to the cpu_reset TLB code

Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux

Pull arm64 fixes from Catalin Marinas:

 - Don't taint the kernel if CPUs have different sets of page sizes
   supported (other than the one in use).

 - Issue I-cache maintenance for module ftrace trampoline.

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: ftrace: Ensure module ftrace trampoline is coherent with I-side
  arm64: cpufeature: Don't treat granule sizes as strict

arm64: ftrace: Ensure module ftrace trampoline is coherent with I-side

The initial support for dynamic ftrace trampolines in modules made use
of an indirect branch which loaded its target from the beginning of
a special section (e71a4e1bebaf7 ("arm64: ftrace: add support for far
branches to dynamic ftrace")). Since no instructions were being patched,
no cache maintenance was needed. However, later in be0f272bfc83 ("arm64:
ftrace: emit ftrace-mod.o contents through code") this code was reworked
to output the trampoline instructions directly into the PLT entry but,
unfortunately, the necessary cache maintenance was overlooked.

Add a call to __flush_icache_range() after writing the new trampoline
instructions but before patching in the branch to the trampoline.

Cc: Ard Biesheuvel <[email protected]>
Cc: James Morse <[email protected]>
Cc: <[email protected]>
Fixes: be0f272bfc83 ("arm64: ftrace: emit ftrace-mod.o contents through code")
Signed-off-by: Will Deacon <[email protected]>
Signed-off-by: Catalin Marinas <[email protected]>

Merge tag 'pm-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm

Pull power management fixes from Rafael Wysocki:
 "These add a check to avoid recent suspend-to-idle power regression on
  systems with NVMe drives where the PCIe ASPM policy is "performance"
  (or when the kernel is built without ASPM support), fix an issue
  related to frequency limits in the schedutil cpufreq governor and fix
  a mistake related to the PM QoS usage in the cpufreq core introduced
  recently.

  Specifics:

   - Disable NVMe power optimization related to suspend-to-idle added
     recently on systems where PCIe ASPM is not able to put PCIe links
     into low-power states to prevent excess power from being drawn by
     the system while suspended (Rafael Wysocki).

   - Make the schedutil governor handle frequency limits changes
     properly in all cases (Viresh Kumar).

   - Prevent the cpufreq core from treating positive values returned by
     dev_pm_qos_update_request() as errors (Viresh Kumar)"

* tag 'pm-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  nvme-pci: Allow PCI bus-level PM to be used if ASPM is disabled
  PCI/ASPM: Add pcie_aspm_enabled()
  cpufreq: schedutil: Don't skip freq update when limits change
  cpufreq: dev_pm_qos_update_request() can return 1 on success

Merge tag 'dmaengine-fix-5.3-rc5' of git://git.infradead.org/users/vkoul/slave-dma

Pull dmaengine fixes from Vinod Koul:
 "Fixes in dmaengine drivers for:

   - dw-edma: endianess, _iomem type and stack usages

   - ste_dma40: unneeded variable and null-pointer dereference

   - tegra210-adma: unused function

   - omap-dma: off-by-one fix"

* tag 'dmaengine-fix-5.3-rc5' of git://git.infradead.org/users/vkoul/slave-dma:
  omap-dma/omap_vout_vrfb: fix off-by-one fi value
  dmaengine: stm32-mdma: Fix a possible null-pointer dereference in stm32_mdma_irq_handler()
  dmaengine: tegra210-adma: Fix unused function warnings
  dmaengine: ste_dma40: fix unneeded variable warning
  dmaengine: dw-edma: fix endianess confusion
  dmaengine: dw-edma: fix __iomem type confusion
  dmaengine: dw-edma: fix unnecessary stack usage

Merge tag 'sound-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "All small fixes targeted for stable:

   - Two fixes for USB-audio with malformed descriptor, spotted by
     fuzzers

   - Two fixes Conexant HD-audio codec wrt power management

   - Quirks for HD-audio AMD platform and HP laptop

   - HD-audio memory leak fix"

* tag 'sound-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: usb-audio: Fix a stack buffer overflow bug in check_input_term
  ALSA: usb-audio: Fix an OOB bug in parse_audio_mixer_unit
  ALSA: hda - Add a generic reboot_notify
  ALSA: hda - Let all conexant codec enter D3 when rebooting
  ALSA: hda/realtek - Add quirk for HP Envy x360
  ALSA: hda - Fix a memory leak bug
  ALSA: hda - Apply workaround for another AMD chip 1022:1487

Merge tag 'drm-fixes-2019-08-16' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "Nothing too crazy this week, one amdgpu fix to use vmalloc for a
  struct that grew in size, and another MST fix for nouveau, and some
  other misc fixes:

  i915:
   - single GVT use after free fix

  scheduler:
   - entity destruction race fix

  amdgpu:
   - struct allocation fix
   - gfx9 soft recovery fix

  nouveau:
   - followup MST fix

  ast:
   - vga register race fix"

* tag 'drm-fixes-2019-08-16' of git://anongit.freedesktop.org/drm/drm:
  drm/nouveau: Only recalculate PBN/VCPI on mode/connector changes
  drm/ast: Fixed reboot test may cause system hanged
  drm/scheduler: use job count instead of peek
  drm/amd/display: use kvmalloc for dc_state (v2)
  drm/amdgpu: fix gfx9 soft recovery
  drm/i915: Use after free in error path in intel_vgpu_create_workload()

Merge branch 'pm-cpufreq'

* pm-cpufreq:
  cpufreq: schedutil: Don't skip freq update when limits change
  cpufreq: dev_pm_qos_update_request() can return 1 on success

x86/boot: Save fields explicitly, zero out everything else

Recent gcc compilers (gcc 9.1) generate warnings about an out of bounds
memset, if the memset goes accross several fields of a struct. This
generated a couple of warnings on x86_64 builds in sanitize_boot_params().

Fix this by explicitly saving the fields in struct boot_params
that are intended to be preserved, and zeroing all the rest.

[ tglx: Tagged for stable as it breaks the warning free build there as well ]

Suggested-by: Thomas Gleixner <[email protected]>
Suggested-by: H. Peter Anvin <[email protected]>
Signed-off-by: John Hubbard <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]

Merge tag 'soundwire-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire into char-misc-linus

Vinod writes:

soundwire fixes for v5.3-rc5

Pierre sent fixes which are queued now for v5.3-rc5 are:
 - regmap dependecy
 - cadence register definitions

* tag 'soundwire-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire:
  soundwire: fix regmap dependencies and align with other serial links
  soundwire: cadence_master: fix definitions for INTSTAT0/1
  soundwire: cadence_master: fix register definition for SLAVE_STATE

Merge tag 'drm-intel-fixes-2019-08-15' of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes

drm/i915 fixes for v5.4-rc5:
- GVT use-after-free fix

Signed-off-by: Dave Airlie <[email protected]>
From: Jani Nikula <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

ALSA: usb-audio: Fix a stack buffer overflow bug in check_input_term

`check_input_term` recursively calls itself with input from
device side (e.g., uac_input_terminal_descriptor.bCSourceID)
as argument (id). In `check_input_term`, if `check_input_term`
is called with the same `id` argument as the caller, it triggers
endless recursive call, resulting kernel space stack overflow.

This patch fixes the bug by adding a bitmap to `struct mixer_build`
to keep track of the checked ids and stop the execution if some id
has been checked (similar to how parse_audio_unit handles unitid
argument).

Reported-by: Hui Peng <[email protected]>
Reported-by: Mathias Payer <[email protected]>
Signed-off-by: Hui Peng <[email protected]>
Cc: <[email protected]>
Signed-off-by: Takashi Iwai <[email protected]>

Merge tag 'xfs-5.3-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull xfs fixes from Darrick Wong:

 - Fix crashes when the attr fork isn't present due to errors but inode
   inactivation tries to zap the attr data anyway.

 - Convert more directory corruption debugging asserts to actual
   EFSCORRUPTED returns instead of blowing up later on.

 - Don't fail writeback just because we ran out of memory allocating
   metadata log data.

* tag 'xfs-5.3-fixes-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: don't crash on null attr fork xfs_bmapi_read
  xfs: remove more ondisk directory corruption asserts
  fs: xfs: xfs_log: Don't use KM_MAYFAIL at xfs_log_reserve().

Merge tag 'iomap-5.3-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull iomap fixlet from Darrick Wong:
 "A single update to the MAINTAINERS entry for iomap now that we've
  removed fs/iomap.c"

* tag 'iomap-5.3-fixes-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  MAINTAINERS: iomap: Remove fs/iomap.c record

io_uring: fix an issue when IOSQE_IO_LINK is inserted into defer list

This patch may fix two issues:

First, when IOSQE_IO_DRAIN set, the next IOs need to be inserted into
defer list to delay execution, but link io will be actively scheduled to
run by calling io_queue_sqe.

Second, when multiple LINK_IOs are inserted together with defer_list,
the LINK_IO is no longer keep order.

   |-------------|
   |   LINK_IO   |      ----> insert to defer_list  -----------
   |-------------|                                            |
   |   LINK_IO   |      ----> insert to defer_list  ----------|
   |-------------|                                            |
   |   LINK_IO   |      ----> insert to defer_list  ----------|
   |-------------|                                            |
   |   NORMAL_IO |      ----> insert to defer_list  ----------|
   |-------------|                                            |
                                                              |
                              queue_work at same time   <-----|

Fixes: 9e645e1105c ("io_uring: add support for sqe links")
Signed-off-by: Jackie Liu <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>

block: remove REQ_NOWAIT_INLINE

We had a few issues with this code, and there's still a problem around
how we deal with error handling for chained/split bios. For now, just
revert the code and we'll try again with a thoroug solution. This
reverts commits:

e15c2ffa1091 ("block: fix O_DIRECT error handling for bio fragments")
0eb6ddfb865c ("block: Fix __blkdev_direct_IO() for bio fragments")
6a43074e2f46 ("block: properly handle IOCB_NOWAIT for async O_DIRECT IO")
893a1c97205a ("blk-mq: allow REQ_NOWAIT to return an error inline")

Signed-off-by: Jens Axboe <[email protected]>

io_uring: fix manual setup of iov_iter for fixed buffers

Commit bd11b3a391e3 ("io_uring: don't use iov_iter_advance() for fixed
buffers") introduced an optimization to avoid using the slow
iov_iter_advance by manually populating the iov_iter iterator in some
cases.

However, the computation of the iterator count field was erroneous: The
first bvec was always accounted for an extent of page size even if the
bvec length was smaller.

In consequence, some I/O operations on fixed buffers were unable to
operate on the full extent of the buffer, consistently skipping some
bytes at the end of it.

Fixes: bd11b3a391e3 ("io_uring: don't use iov_iter_advance() for fixed buffers")
Cc: [email protected]
Signed-off-by: Aleix Roca Nonell <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>

Merge tag 'auxdisplay-for-linus-v5.3-rc5' of git://github.com/ojeda/linux

Pull auxdisplay fixes from Miguel Ojeda:
 "A few minor auxdisplay improvements:

   - A couple of small header cleanups for charlcd (Masahiro Yamada)

   - A trivial typo fix for the examples of cfag12864b (Masahiro Yamada)

   - An Kconfig help text improvement for charlcd (Mans Rullgard)

   - An error path fix for panel (zhengbin)"

* tag 'auxdisplay-for-linus-v5.3-rc5' of git://github.com/ojeda/linux:
  auxdisplay: Fix a typo in cfag12864b-example.c
  auxdisplay: charlcd: add include guard to charlcd.h
  auxdisplay: charlcd: move charlcd.h to drivers/auxdisplay
  auxdisplay: charlcd: add help text for backlight initial state
  auxdisplay: panel: need to delete scan_timer when misc_register fails in panel_attach

Merge tag 'devicetree-fixes-for-5.3-3' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux

Pull devicetree fixes from Rob Herring:

 - Fix building DT binding examples for in tree builds

 - Correct some refcounting in adjust_local_phandle_references()

 - Update FSL FEC binding with deprecated properties

 - Schema fix in stm32 pinctrl

 - Fix typo in of_irq_parse_one docbook comment

* tag 'devicetree-fixes-for-5.3-3' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux:
  of: irq: fix a trivial typo in a doc comment
  dt-bindings: pinctrl: stm32: Fix 'st,syscfg' schema
  dt-bindings: fec: explicitly mark deprecated properties
  of: resolver: Add of_node_put() before return and break
  dt-bindings: Fix generated example files getting added to schemas

misc: xilinx-sdfec: fix dependency and build error

lib/devres.c, which implements devm_ioremap_resource(), is only built
when CONFIG_HAS_IOMEM is set/enabled, so XILINX_SDFEC should depend
on HAS_IOMEM.  Fixes this build error (as seen on UML builds):

ERROR: "devm_ioremap_resource" [drivers/misc/xilinx_sdfec.ko] undefined!

Fixes: 76d83e1c3233 ("misc: xilinx-sdfec: add core driver")
Signed-off-by: Randy Dunlap <[email protected]>
Cc: Derek Kiernan <[email protected]>
Cc: Dragan Cvetic <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: Geert Uytterhoeven <[email protected]>
Cc: Arnd Bergmann <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

usb: add a hcd_uses_dma helper

The USB buffer allocation code is the only place in the usb core (and in
fact the whole kernel) that uses is_device_dma_capable, while the URB
mapping code uses the uses_dma flag in struct usb_bus.  Switch the buffer
allocation to use the uses_dma flag used by the rest of the USB code,
and create a helper in hcd.h that checks this flag as well as the
CONFIG_HAS_DMA to simplify the caller a bit.

Signed-off-by: Christoph Hellwig <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

usb: don't create dma pools for HCDs with a localmem_pool

If the HCD provides a localmem pool we will never use the DMA pools, so
don't create them.

Fixes: b0310c2f09bb ("USB: use genalloc for USB HCs with local memory")
Signed-off-by: Christoph Hellwig <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

usb: chipidea: imx: fix EPROBE_DEFER support during driver probe

If driver probe needs to be deferred, e.g. because ci_hdrc_add_device()
isn't ready yet, this driver currently misbehaves badly:
    a) success is still reported to the driver core (meaning a 2nd
       probe attempt will never be done), leaving the driver in
       a dysfunctional state and the hardware unusable

    b) driver remove / shutdown OOPSes:
    [  206.786916] Unable to handle kernel paging request at virtual address fffffdff
    [  206.794148] pgd = 880b9f82
    [  206.796890] [fffffdff] *pgd=abf5e861, *pte=00000000, *ppte=00000000
    [  206.803179] Internal error: Oops: 37 [#1] PREEMPT SMP ARM
    [  206.808581] Modules linked in: wl18xx evbug
    [  206.813308] CPU: 1 PID: 1 Comm: systemd-shutdow Not tainted 4.19.35+gf345c93b4195 #1
    [  206.821053] Hardware name: Freescale i.MX7 Dual (Device Tree)
    [  206.826813] PC is at ci_hdrc_remove_device+0x4/0x20
    [  206.831699] LR is at ci_hdrc_imx_remove+0x20/0xe8
    [  206.836407] pc : [<805cd4b0>]    lr : [<805d62cc>]    psr: 20000013
    [  206.842678] sp : a806be40  ip : 00000001  fp : 80adbd3c
    [  206.847906] r10: 80b1b794  r9 : 80d5dfe0  r8 : a8192c44
    [  206.853136] r7 : 80db93a0  r6 : a8192c10  r5 : a8192c00  r4 : a93a4a00
    [  206.859668] r3 : 00000000  r2 : a8192ce4  r1 : ffffffff  r0 : fffffdfb
    [  206.866201] Flags: nzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment none
    [  206.873341] Control: 10c5387d  Table: a9e0c06a  DAC: 00000051
    [  206.879092] Process systemd-shutdow (pid: 1, stack limit = 0xb271353c)
    [  206.885624] Stack: (0xa806be40 to 0xa806c000)
    [  206.889992] be40: a93a4a00 805d62cc a8192c1c a8170e10 a8192c10 8049a490 80d04d08 00000000
    [  206.898179] be60: 00000000 80d0da2c fee1dead 00000000 a806a000 00000058 00000000 80148b08
    [  206.906366] be80: 01234567 80148d8c a9858600 00000000 00000000 00000000 00000000 80d04d08
    [  206.914553] bea0: 00000000 00000000 a82741e0 a9858600 00000024 00000002 a9858608 00000005
    [  206.922740] bec0: 0000001e 8022c058 00000000 00000000 a806bf14 a9858600 00000000 a806befc
    [  206.930927] bee0: a806bf78 00000000 7ee12c30 8022c18c a806bef8 a806befc 00000000 00000001
    [  206.939115] bf00: 00000000 00000024 a806bf14 00000005 7ee13b34 7ee12c68 00000004 7ee13f20
    [  206.947302] bf20: 00000010 7ee12c7c 00000005 7ee12d04 0000000a 76e7dc00 00000001 80d0f140
    [  206.955490] bf40: ab637880 a974de40 60000013 80d0f140 ab6378a0 80d04d08 a8080470 a9858600
    [  206.963677] bf60: a9858600 00000000 00000000 8022c24c 00000000 80144310 00000000 00000000
    [  206.971864] bf80: 80101204 80d04d08 00000000 80d04d08 00000000 00000000 00000003 00000058
    [  206.980051] bfa0: 80101204 80101000 00000000 00000000 fee1dead 28121969 01234567 00000000
    [  206.988237] bfc0: 00000000 00000000 00000003 00000058 00000000 00000000 00000000 00000000
    [  206.996425] bfe0: 0049ffb0 7ee13d58 0048a84b 76f245a6 60000030 fee1dead 00000000 00000000
    [  207.004622] [<805cd4b0>] (ci_hdrc_remove_device) from [<805d62cc>] (ci_hdrc_imx_remove+0x20/0xe8)
    [  207.013509] [<805d62cc>] (ci_hdrc_imx_remove) from [<8049a490>] (device_shutdown+0x16c/0x218)
    [  207.022050] [<8049a490>] (device_shutdown) from [<80148b08>] (kernel_restart+0xc/0x50)
    [  207.029980] [<80148b08>] (kernel_restart) from [<80148d8c>] (sys_reboot+0xf4/0x1f0)
    [  207.037648] [<80148d8c>] (sys_reboot) from [<80101000>] (ret_fast_syscall+0x0/0x54)
    [  207.045308] Exception stack(0xa806bfa8 to 0xa806bff0)
    [  207.050368] bfa0:                   00000000 00000000 fee1dead 28121969 01234567 00000000
    [  207.058554] bfc0: 00000000 00000000 00000003 00000058 00000000 00000000 00000000 00000000
    [  207.066737] bfe0: 0049ffb0 7ee13d58 0048a84b 76f245a6
    [  207.071799] Code: ebffffa8 e3a00000 e8bd8010 e92d4010 (e5904004)
    [  207.078021] ---[ end trace be47424e3fd46e9f ]---
    [  207.082647] Kernel panic - not syncing: Fatal exception
    [  207.087894] ---[ end Kernel panic - not syncing: Fatal exception ]---

    c) the error path in combination with driver removal causes
       imbalanced calls to the clk_*() and pm_()* APIs

a) happens because the original intended return value is
   overwritten (with 0) by the return code of
   regulator_disable() in ci_hdrc_imx_probe()'s error path
b) happens because ci_pdev is -EPROBE_DEFER, which causes
   ci_hdrc_remove_device() to OOPS

Fix a) by being more careful in ci_hdrc_imx_probe()'s error
path and not overwriting the real error code

Fix b) by calling the respective cleanup functions during
remove only when needed (when ci_pdev != NULL, i.e. when
everything was initialised correctly). This also has the
side effect of not causing imbalanced clk_*() and pm_*()
API calls as part of the error code path.

Fixes: 7c8e8909417e ("usb: chipidea: imx: add HSIC support")
Signed-off-by: André Draszik <[email protected]>
Cc: stable <[email protected]>
CC: Peter Chen <[email protected]>
CC: Greg Kroah-Hartman <[email protected]>
CC: Shawn Guo <[email protected]>
CC: Sascha Hauer <[email protected]>
CC: Pengutronix Kernel Team <[email protected]>
CC: Fabio Estevam <[email protected]>
CC: NXP Linux Team <[email protected]>
CC: [email protected]
CC: [email protected]
CC: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

usb: host: fotg2: restart hcd after port reset

On the Gemini SoC the FOTG2 stalls after port reset
so restart the HCD after each port reset.

Signed-off-by: Hans Ulli Kroll <[email protected]>
Signed-off-by: Linus Walleij <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

USB: CDC: fix sanity checks in CDC union parser

A few checks checked for the size of the pointer to a structure
instead of the structure itself. Copy & paste issue presumably.

Fixes: e4c6fb7794982 ("usbnet: move the CDC parser into USB core")
Cc: stable <[email protected]>
Reported-by: [email protected]
Signed-off-by: Oliver Neukum <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

usb: cdc-acm: make sure a refcount is taken early enough

destroy() will decrement the refcount on the interface, so that
it needs to be taken so early that it never undercounts.

Fixes: 7fb57a019f94e ("USB: cdc-acm: Fix potential deadlock (lockdep warning)")
Cc: stable <[email protected]>
Reported-and-tested-by: [email protected]
Signed-off-by: Oliver Neukum <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: sm750fb: fix odd license text

There is some "confusing" license text in some of the sm750fb driver
files.  After discussing it with the company, it turns out to have been
a mistake and these lines can be safely removed.  The files are all to
be licensed under the GPLv2 license.

Cc: Sudip Mukherjee <[email protected]>
Cc: Teddy Wang <[email protected]>
Cc: [email protected]
Cc: [email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: correct all misused ENOTSUPP

As Chao pointed out [1], ENOTSUPP is used for NFS
protocol only, we should use EOPNOTSUPP instead...

[1] https://lore.kernel.org/lkml/108ee2f9-75dd-b8ab-8da7-b81c17bafbf6@huawei.com/

Reported-by: Chao Yu <[email protected]>
Signed-off-by: Gao Xiang <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: differentiate unsupported on-disk format

For some specific fields, use EOPNOTSUPP instead of EIO
for values which look sane but aren't supported right now.

Reviewed-by: Chao Yu <[email protected]>
Signed-off-by: Gao Xiang <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: introduce EFSCORRUPTED and more logs

Previously, EROFS uses EIO to indicate that filesystem
is corrupted as well. However, as Pavel said [1], other
filesystems tend to use EUCLEAN(EFSCORRUPTED) instead,
let's follow what others do right now.

Also, add some more prints to the syslog.

[1] https://lore.kernel.org/lkml/20190813114821.GB11559@amd/

Suggested-by: Pavel Machek <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Signed-off-by: Gao Xiang <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: comedi: usbduxsigma: remove redundant assignment to variable fx2delay

Variable fx2delay is being initialized with a value that is never read
and fx2delay is being re-assigned a little later on. The assignment is
redundant and hence can be removed.

Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: rtl8192u: fix up indentation of a statement

There is a statement that is indented one level too deeply, remove
the extraneous tab.

Addresses-Coverity: ("Identation does not match nesting level")
Signed-off-by: Colin Ian King <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: gasket: apex: Make structure apex_desc constant

Static structure apex_desc, of type gasket_driver_desc, is used only as
an argument to the functions gasket_register_device() and
gasket_unregister_device(). In the definitions of both these functions,
their parameter is declared as const. Hence make apex_desc itself
constant to protect it from modification.
Issue found with Coccinelle.

Signed-off-by: Nishka Dasgupta <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

staging: erofs: removing an extra call to iloc() in fill_inode()

in fill_inode() we call iloc() twice.Avoiding the extra call by
storing the result.

Signed-off-by: Pratik Shinde <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Cc: Gao Xiang <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

USB: serial: option: add the BroadMobi BM818 card

Add a VID:PID for the BroadMobi BM818 M.2 card

T:  Bus=01 Lev=03 Prnt=40 Port=03 Cnt=01 Dev#= 44 Spd=480 MxCh= 0
D:  Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
P:  Vendor=2020 ProdID=2060 Rev=00.00
S:  Manufacturer=Qualcomm, Incorporated
S:  Product=Qualcomm CDMA Technologies MSM
C:  #Ifs= 5 Cfg#= 1 Atr=e0 MxPwr=500mA
I:  If#=0x0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
I:  If#=0x1 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
I:  If#=0x2 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
I:  If#=0x3 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=fe Prot=ff Driver=(none)
I:  If#=0x4 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)

Signed-off-by: Bob Ham <[email protected]>
Signed-off-by: Angus Ainslie (Purism) <[email protected]>
Cc: stable <[email protected]>
[ johan: use USB_DEVICE_INTERFACE_CLASS() ]
Signed-off-by: Johan Hovold <[email protected]>

USB: serial: option: Add Motorola modem UARTs

On Motorola Mapphone devices such as Droid 4 there are five USB ports
that do not use the same layout as Gobi 1K/2K/etc devices listed in
qcserial.c. So we should use qcaux.c or option.c as noted by
Dan Williams <[email protected]>.

As the Motorola USB serial ports have an interrupt endpoint as shown
with lsusb -v, we should use option.c instead of qcaux.c as pointed out
by Johan Hovold <[email protected]>.

The ff/ff/ff interfaces seem to always be UARTs on Motorola devices.
For the other interfaces, class 0x0a (CDC Data) should not in general
be added as they are typically part of a multi-interface function as
noted earlier by Bjørn Mork <[email protected]>.

However, looking at the Motorola mapphone kernel code, the mdm6600 0x0a
class is only used for flashing the modem firmware, and there are no
other interfaces. So I've added that too with more details below as it
works just fine.

The ttyUSB ports on Droid 4 are:

ttyUSB0 DIAG, CQDM-capable
ttyUSB1 MUX or NMEA, no response
ttyUSB2 MUX or NMEA, no response
ttyUSB3 TCMD
ttyUSB4 AT-capable

The ttyUSB0 is detected as QCDM capable by ModemManager. I think
it's only used for debugging with ModemManager --debug for sending
custom AT commands though. ModemManager already can manage data
connection using the USB QMI ports that are already handled by the
qmi_wwan.c driver.

To enable the MUX or NMEA ports, it seems that something needs to be
done additionally to enable them, maybe via the DIAG or TCMD port.
It might be just a NVRAM setting somewhere, but I have no idea what
NVRAM settings may need changing for that.

The TCMD port seems to be a Motorola custom protocol for testing
the modem and to configure it's NVRAM and seems to work just fine
based on a quick test with a minimal tcmdrw tool I wrote.

The voice modem AT-capable port seems to provide only partial
support, and no PM support compared to the TS 27.010 based UART
wired directly to the modem.

The UARTs added with this change are the same product IDs as the
Motorola Mapphone Android Linux kernel mdm6600_id_table. I don't
have any mdm9600 based devices, so I have only tested these on
mdm6600 based droid 4.

Then for the class 0x0a (CDC Data) mode, the Motorola Mapphone Android
Linux kernel driver moto_flashqsc.c just seems to change the
port->bulk_out_size to 8K from the default. And is only used for
flashing the modem firmware it seems.

I've verified that flashing the modem with signed firmware works just
fine with the option driver after manually toggling the GPIO pins, so
I've added droid 4 modem flashing mode to the option driver. I've not
added the other devices listed in moto_flashqsc.c in case they really
need different port->bulk_out_size. Those can be added as they get
tested to work for flashing the modem.

After this patch the output of /sys/kernel/debug/usb/devices has
the following for normal 22b8:2a70 mode including the related qmi_wwan
interfaces:

T:  Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#=  2 Spd=12   MxCh= 0
D:  Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
P:  Vendor=22b8 ProdID=2a70 Rev= 0.00
S:  Manufacturer=Motorola, Incorporated
S:  Product=Flash MZ600
C:* #Ifs= 9 Cfg#= 1 Atr=e0 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
E:  Ad=81(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=01(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 2 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
E:  Ad=83(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=03(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 3 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
E:  Ad=84(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=04(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 4 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
E:  Ad=85(I) Atr=03(Int.) MxPS=  64 Ivl=5ms
E:  Ad=86(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=05(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 5 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=fb Prot=ff Driver=qmi_wwan
E:  Ad=87(I) Atr=03(Int.) MxPS=  64 Ivl=5ms
E:  Ad=88(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=06(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 6 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=fb Prot=ff Driver=qmi_wwan
E:  Ad=89(I) Atr=03(Int.) MxPS=  64 Ivl=5ms
E:  Ad=8a(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=07(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 7 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=fb Prot=ff Driver=qmi_wwan
E:  Ad=8b(I) Atr=03(Int.) MxPS=  64 Ivl=5ms
E:  Ad=8c(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=08(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 8 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=fb Prot=ff Driver=qmi_wwan
E:  Ad=8d(I) Atr=03(Int.) MxPS=  64 Ivl=5ms
E:  Ad=8e(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=09(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms

In 22b8:900e "qc_dload" mode the device shows up as:

T:  Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#=  2 Spd=12   MxCh= 0
D:  Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
P:  Vendor=22b8 ProdID=900e Rev= 0.00
S:  Manufacturer=Motorola, Incorporated
S:  Product=Flash MZ600
C:* #Ifs= 1 Cfg#= 1 Atr=e0 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=option
E:  Ad=81(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=01(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms

And in 22b8:4281 "ram_downloader" mode the device shows up as:

T:  Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#=  2 Spd=12   MxCh= 0
D:  Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
P:  Vendor=22b8 ProdID=4281 Rev= 0.00
S:  Manufacturer=Motorola, Incorporated
S:  Product=Flash MZ600
C:* #Ifs= 1 Cfg#= 1 Atr=e0 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 2 Cls=0a(data ) Sub=00 Prot=fc Driver=option
E:  Ad=81(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=01(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms

Cc: Bjørn Mork <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: Lars Melin <[email protected]>
Cc: Marcel Partap <[email protected]>
Cc: Merlijn Wajer <[email protected]>
Cc: Michael Scott <[email protected]>
Cc: NeKit <[email protected]>
Cc: Pavel Machek <[email protected]>
Cc: Sebastian Reichel <[email protected]>
Tested-by: Pavel Machek <[email protected]>
Signed-off-by: Tony Lindgren <[email protected]>
Cc: stable <[email protected]>
Signed-off-by: Johan Hovold <[email protected]>

MAINTAINERS, x86/CPU: Tony Luck will maintain asm/intel-family.h

There are a few different subsystems in the kernel that depend on model
specific behaviour (perf, EDAC, power, ...). Easier for just one person
to have the task to get new model numbers included instead of having
these groups trip over each other to do it.

 [ bp: s/Cpu/CPU/ and add [email protected] so that it gets CCed too as
   FYI. ]

Signed-off-by: Tony Luck <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Acked-by: Thomas Gleixner <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: x86-ml <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]

Merge tag 'drm-fixes-5.3-2019-08-14' of git://people.freedesktop.org/~agd5f/linux into drm-fixes

drm-fixes-5.3-2019-08-14:

amdgpu:
- Use kvalloc for dc_state to avoid allocation
  failures in some cases.
- Fix gfx9 soft recovery

scheduler:
- Fix a race condition when destroying entities

Signed-off-by: Dave Airlie <[email protected]>
From: Alex Deucher <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/nouveau: Only recalculate PBN/VCPI on mode/connector changes

I -thought- I had fixed this entirely, but it looks like that I didn't
test this thoroughly enough as we apparently still make one big mistake
with nv50_msto_atomic_check() - we don't handle the following scenario:

* CRTC #1 has n VCPI allocated to it, is attached to connector DP-4
  which is attached to encoder #1. enabled=y active=n
* CRTC #1 is changed from DP-4 to DP-5, causing:
  * DP-4 crtc=#1→NULL (VCPI n→0)
  * DP-5 crtc=NULL→#1
  * CRTC #1 steals encoder #1 back from DP-4 and gives it to DP-5
  * CRTC #1 maintains the same mode as before, just with a different
    connector
* mode_changed=n connectors_changed=y
  (we _SHOULD_ do VCPI 0→n here, but don't)

Once the above scenario is repeated once, we'll attempt freeing VCPI
from the connector that we didn't allocate due to the connectors
changing, but the mode staying the same. Sigh.

Since nv50_msto_atomic_check() has broken a few times now, let's rethink
things a bit to be more careful: limit both VCPI/PBN allocations to
mode_changed || connectors_changed, since neither VCPI or PBN should
ever need to change outside of routing and mode changes.

Changes since v1:
* Fix accidental reversal of clock and bpp arguments in
  drm_dp_calc_pbn_mode() - William Lewis

Signed-off-by: Lyude Paul <[email protected]>
Reported-by: Bohdan Milar <[email protected]>
Tested-by: Bohdan Milar <[email protected]>
Fixes: 232c9eec417a ("drm/nouveau: Use atomic VCPI helpers for MST")
References: 412e85b60531 ("drm/nouveau: Only release VCPI slots on mode changes")
Cc: Lyude Paul <[email protected]>
Cc: Ben Skeggs <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: David Airlie <[email protected]>
Cc: Jerry Zuo <[email protected]>
Cc: Harry Wentland <[email protected]>
Cc: Juston Li <[email protected]>
Cc: Laurent Pinchart <[email protected]>
Cc: Karol Herbst <[email protected]>
Cc: Ilia Mirkin <[email protected]>
Cc: <[email protected]> # v5.1+
Acked-by: Ben Skeggs <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

drm/ast: Fixed reboot test may cause system hanged

There is another thread still access standard VGA I/O while loading drm driver.
Disable standard VGA I/O decode to avoid this issue.

Signed-off-by: Y.C. Chen <[email protected]>
Reviewed-by: Benjamin Herrenschmidt <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

of: irq: fix a trivial typo in a doc comment

Diverged from what the code does with commit 530210c7814e ("of/irq: Replace
of_irq with of_phandle_args").

Signed-off-by: Lubomir Rintel <[email protected]>
Signed-off-by: Rob Herring <[email protected]>

dt-bindings: pinctrl: stm32: Fix 'st,syscfg' schema

The proper way to add additional contraints to an existing json-schema
is using 'allOf' to reference the base schema. Using just '$ref' doesn't
work. Fix this for the 'st,syscfg' property.

Cc: Mark Rutland <[email protected]>
Cc: Maxime Coquelin <[email protected]>
Cc: Alexandre Torgue <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Reviewed-by: Linus Walleij <[email protected]>
Signed-off-by: Rob Herring <[email protected]>

Merge tag 'Wimplicit-fallthrough-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gustavoars/linux

Pull fallthrough fixes from Gustavo A. R. Silva:
 "Fix sh mainline builds:

   - Fix fall-through warning in sh.

   - Fix missing break bug in sh (this is a 10-year-old bug)

  Currently, mainline builds for sh are broken. These patches fix that"

* tag 'Wimplicit-fallthrough-5.3-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gustavoars/linux:
  sh: kernel: hw_breakpoint: Fix missing break in switch statement
  sh: kernel: disassemble: Mark expected switch fall-throughs

Merge tag 'afs-fixes-20190814' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs

Pull afs fixes from David Howells:

 - Fix the CB.ProbeUuid handler to generate its reply correctly.

 - Fix a mix up in indices when parsing a Volume Location entry record.

 - Fix a potential NULL-pointer deref when cleaning up a read request.

 - Fix the expected data version of the destination directory in
   afs_rename().

 - Fix afs_d_revalidate() to only update d_fsdata if it's not the same
   as the directory data version to reduce the likelihood of overwriting
   the result of a competing operation. (d_fsdata carries the directory
   DV or the least-significant word thereof).

 - Fix the tracking of the data-version on a directory and make sure
   that dentry objects get properly initialised, updated and
   revalidated.

   Also fix rename to update d_fsdata to match the new directory's DV if
   the dentry gets moved over and unhash the dentry to stop
   afs_d_revalidate() from interfering.

* tag 'afs-fixes-20190814' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs:
  afs: Fix missing dentry data version updating
  afs: Only update d_fsdata if different in afs_d_revalidate()
  afs: Fix off-by-one in afs_rename() expected data version calculation
  fs: afs: Fix a possible null-pointer dereference in afs_put_read()
  afs: Fix loop index mixup in afs_deliver_vl_get_entry_by_name_u()
  afs: Fix the CB.ProbeUuid service handler to reply correctly

drm/scheduler: use job count instead of peek

The spsc_queue_peek function is accessing queue->head which belongs to
the consumer thread and shouldn't be accessed by the producer

This is fixing a rare race condition when destroying entities.

Signed-off-by: Christian König <[email protected]>
Acked-by: Andrey Grodzovsky <[email protected]>
Reviewed-by: [email protected]
Signed-off-by: Alex Deucher <[email protected]>

riscv: Make __fstate_clean() work correctly.

Make the __fstate_clean() function correctly set the
state of sstatus.FS in pt_regs to SR_FS_CLEAN.

Fixes: 7db91e57a0acd ("RISC-V: Task implementation")
Cc: linux-stable <[email protected]>
Signed-off-by: Vincent Chen <[email protected]>
Reviewed-by: Anup Patel <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
[[email protected]: expanded "Fixes" commit ID]
Signed-off-by: Paul Walmsley <[email protected]>

riscv: Correct the initialized flow of FP register

  The following two reasons cause FP registers are sometimes not
initialized before starting the user program.
1. Currently, the FP context is initialized in flush_thread() function
   and we expect these initial values to be restored to FP register when
   doing FP context switch. However, the FP context switch only occurs in
   switch_to function. Hence, if this process does not be scheduled out
   and scheduled in before entering the user space, the FP registers
   have no chance to initialize.
2. In flush_thread(), the state of reg->sstatus.FS inherits from the
   parent. Hence, the state of reg->sstatus.FS may be dirty. If this
   process is scheduled out during flush_thread() and initializing the
   FP register, the fstate_save() in switch_to will corrupt the FP context
   which has been initialized until flush_thread().

  To solve the 1st case, the initialization of the FP register will be
completed in start_thread(). It makes sure all FP registers are initialized
before starting the user program. For the 2nd case, the state of
reg->sstatus.FS in start_thread will be set to SR_FS_OFF to prevent this
process from corrupting FP context in doing context save. The FP state is
set to SR_FS_INITIAL in start_trhead().

Signed-off-by: Vincent Chen <[email protected]>
Reviewed-by: Anup Patel <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Fixes: 7db91e57a0acd ("RISC-V: Task implementation")
Cc: [email protected]
[[email protected]: fixed brace alignment issue reported by
 checkpatch]
Signed-off-by: Paul Walmsley <[email protected]>