Re: [PATCH 5.15.y 0/9] can: bcm: backport of 9 patches that failed to apply to 5.15-stable

[Greg KH <gregkh@linuxfoundation.org>]

On Fri, Jul 24, 2026 at 10:37:47AM +0200, Oliver Hartkopp wrote:
> This resubmits "can: bcm: add locking when updating filter and timer
> values" (commit 749179c2e25b95d22499ed29096b3e02d6dfd2b4 upstream),
> which failed to apply to the 5.15-stable tree, together with 8
> further bcm.c patches from the same upstream series that build on it
> and would otherwise fail for the same reason. This is the full set
> of 11 upstream bcm.c patches from 7.2-rc3 minus the 2 already queued
> by the normal stable process for 5.15 ("can: bcm: fix lockless
> bound/ifindex race and silent RX_SETUP failure", "can: bcm: add
> missing rcu list annotations and operations").
> 
> This assumes the 2 already-queued bcm.c patches for 5.15-stable have
> landed by the time this is applied. If not, please let me know and I
> will rebase.

This is a mess, it showed up like this in our inboxes, with no threading
or information about which patch was for what tree at all:

   1 N   Jul 24 Oliver Hartkopp (2.5K) [PATCH 8/9] can: bcm: fix data race on rx_stamp/rx_ifindex in bcm_rx_handler()                                  
   2 N   Jul 24 Oliver Hartkopp (5.9K) [PATCH 7/9] can: bcm: fix stale rx/tx ops after device removal                                                  
   3 N   Jul 24 Oliver Hartkopp (1.4K) [PATCH 6.12.y 0/8] can: bcm: backport of 8 patches that failed to apply to 6.12-stable                          
   4 N   Jul 24 Oliver Hartkopp (1.5K) [PATCH 5.15.y 0/9] can: bcm: backport of 9 patches that failed to apply to 5.15-stable                          
   8 N   Jul 24 Oliver Hartkopp ( 13K) [PATCH 1/8] can: bcm: add locking when updating filter and timer values                                         
   9     Jul 24 Oliver Hartkopp (2.5K) [PATCH 7/8] can: bcm: fix data race on rx_stamp/rx_ifindex in bcm_rx_handler()
  10 N   Jul 24 Oliver Hartkopp (3.5K) [PATCH 4/8] can: bcm: validate frame length in bcm_rx_setup() for RTR replies                                   
  11 N   Jul 24 Oliver Hartkopp (1.4K) [PATCH 6.1.y 0/8] can: bcm: backport of 8 patches that failed to apply to 6.1-stable                            
  12 N   Jul 24 Oliver Hartkopp (6.7K) [PATCH 3/8] can: bcm: extend bcm_tx_lock usage for data and timer updates                                       
  13 N   Jul 24 Oliver Hartkopp ( 13K) [PATCH 1/8] can: bcm: add locking when updating filter and timer values                                         
  14 N   Jul 24 Oliver Hartkopp (5.2K) [PATCH 8/8] can: bcm: track a single source interface for ANYDEV timeout/throttle ops                           
  15 N   Jul 24 Oliver Hartkopp (4.0K) [PATCH 5/8] can: bcm: add missing device refcount for CAN filter removal                                        
  16 N   Jul 24 Oliver Hartkopp (1.5K) [PATCH 5.10.y 0/9] can: bcm: backport of 9 patches that failed to apply to 5.10-stable                          
  18 N   Jul 24 Oliver Hartkopp (5.9K) [PATCH 6/8] can: bcm: fix stale rx/tx ops after device removal                                                  
  19 N   Jul 24 Oliver Hartkopp (5.2K) [PATCH 9/9] can: bcm: track a single source interface for ANYDEV timeout/throttle ops                           
  20 N   Jul 24 Oliver Hartkopp (2.5K) [PATCH 7/8] can: bcm: fix data race on rx_stamp/rx_ifindex in bcm_rx_handler()                                  
  21 N   Jul 24 Oliver Hartkopp (6.5K) [PATCH 2/8] can: bcm: fix CAN frame rx/tx statistics                                                            
  22 N   Jul 24 Oliver Hartkopp (3.5K) [PATCH 4/8] can: bcm: validate frame length in bcm_rx_setup() for RTR replies                                   
  23 N   Jul 24 Oliver Hartkopp ( 13K) [PATCH 2/9] can: bcm: add locking when updating filter and timer values                                         
  24 N   Jul 24 Oliver Hartkopp (5.9K) [PATCH 1/9] can: bcm: defer rx_op deallocation to workqueue to fix thrtimer UAF                                 
  25 N   Jul 24 Oliver Hartkopp (5.9K) [PATCH 7/9] can: bcm: fix stale rx/tx ops after device removal                                                  
  26 N   Jul 24 Oliver Hartkopp (2.5K) [PATCH 8/9] can: bcm: fix data race on rx_stamp/rx_ifindex in bcm_rx_handler()                                  
  27 N   Jul 24 Oliver Hartkopp (5.9K) [PATCH 1/9] can: bcm: defer rx_op deallocation to workqueue to fix thrtimer UAF                                 
  28 N   Jul 24 Oliver Hartkopp (4.0K) [PATCH 6/9] can: bcm: add missing device refcount for CAN filter removal                                        
  29 N   Jul 24 Oliver Hartkopp (6.5K) [PATCH 3/9] can: bcm: fix CAN frame rx/tx statistics                                                            


I'll delete all of these now, and can you use 'git send-email' or some
such tool that properly links emails together so that we have a chance
to be able to apply them?

thanks,

greg k-h