Re: Request for review for 4.4-st78, 4.4-cip113

Hi!

> 
> Here are the manual backports for the upcoming 4.4 kernel release. As always they can be found in https://git.kernel.org/pub/scm/linux/kernel/git/cip/linux-cip.git/log/?h=linux-4.4.y-st-rc
>

For now, I have not checked the batman-adv stuff. Two comments below.

a > 6fc536e081b4e cdrom, scsi: sr: propagate read-only status to block layer via set_disk_ro()
a > a70db8dc8aa2d bcache: fix cached_dev.sb_bio use-after-free and crash
a operation order in error handling does not match initialization > d6dc394717b9d can: mcp251x: add error handling for power enable in open and resume
> 270f518e2aab7 batman-adv: tt: fix negative tt_buff_len
> b9d56f0fa7881 batman-adv: tt: fix negative last_changeset_len
> 2ea84dbb4204d batman-adv: frag: disallow unicast fragment in fragment
> 3075eb7fe5a6a batman-adv: fix fragment reassembly length accounting
> 5c81177e39927 batman-adv: clear current gateway during teardown
a > 516397de58548 net: usb: rtl8150: fix use-after-free in rtl8150_start_xmit()
a introduces stricter parsing > 2c14f681cb8ca netfilter: nf_conntrack_sip: don't use simple_strtoul
a > a177703144795 netfilter: arp_tables: fix IEEE1394 ARP payload parsing
a > 90d4b856cc573 netfilter: conntrack: remove sprintf usage
a > a5e1c4172bf7b sctp: fix OOB write to userspace in sctp_getsockopt_peer_auth_chunks
a just a robustness against fairly interesting device > a4e3e33596f1b HID: usbhid: fix deadlock in hid_post_reset()
a also removes a comment > b2df99543367b ALSA: core: Validate compress device numbers without dynamic minors
> a856ea763a0c8 batman-adv: bla: put backbone reference on failed claim hash insert
> 420d73ca52ee2 batman-adv: bla: only purge non-released claims
> 283a66b993a78 batman-adv: bla: prevent use-after-free when deleting claims
a just a fix for unusual dts > 1f1f946bac592 power: supply: max17042: avoid overflow when determining health
! no dec in munmap? , c/e, will we now have framebuffer refusing resize? > 0bd353e476ab2 fbdev: udlfb: add vm_ops to dlfb_ops_mmap to prevent use-after-free

I don't see how this works. mmap_count is increased in mmap, but I
don't see corresponding decrease in munmap. 

a > e27540595d282 md/raid5: fix soft lockup in retry_aligned_read()
! check_mul_overflow -- is it generic for all types? does sizeof > return size_t?...? > 16b4cbdf426d7 lib/ts_kmp: fix integer overflow in pattern length calculation

+       unsigned int prefix_tbl_len;
+       if (unlikely(check_mul_overflow(len, sizeof(*kmp->prefix_tbl),
+                                       &prefix_tbl_len) ||

AFAICT sizeof returns size_t, so I'd be much if prefix_tbl_len was
size_t, too.

a just clang fortify workaround > f03a0d998b054 ALSA: control: Validate buf_len before strnlen() in snd_ctl_elem_init_enum_names()
a > 7d2c7f0155186 gfs2: Validate i_depth for exhash directories
a > 06c757446f45d rxrpc: reject undecryptable rxkad response tickets
a > f842b70abca24 wifi: brcmfmac: validate bsscfg indices in IF events
> 

Reviewed-by: Pavel Machek <[email protected]>

Best regards,
								Pavel

-----BEGIN PGP SIGNATURE-----

iF0EABECAB0WIQRPfPO7r0eAhk010v0w5/Bqldv68gUCalS/4wAKCRAw5/Bqldv6
8sjRAJ0ej7kqKA2ot6ThLehAzuJ3WrpJ4QCfedqNJIZ4h3biX0RcBhTIOBRdL/c=
=Gqjo
-----END PGP SIGNATURE-----