XPost: linux.debian.bugs.dist 
 From: kevin@xf.ee 
  
 Hello, 
  
 I've identified the first bad commit using git bisect: 
  
 90bfb28d5fa8127a113a140c9791ea0b40ab156a is the first bad commit 
 commit 90bfb28d5fa8127a113a140c9791ea0b40ab156a 
 Author: Jens Axboe  
 Date:   Tue Sep 10 08:57:04 2024 -0600 
  
     io_uring/rw: drop -EOPNOTSUPP check in __io_complete_rw_common() 
  
     A recent change ensured that the necessary -EOPNOTSUPP -> -EAGAIN 
     transformation happens inline on both the reader and writer side, 
     and hence there's no need to check for both of these anymore on 
     the completion handler side. 
  
     Signed-off-by: Jens Axboe  
  
  io_uring/rw.c | 3 +-- 
  1 file changed, 1 insertion(+), 2 deletions(-) 
  
  
 Here is the git bisect log as well: 
  
 git bisect start 
 # status: waiting for both good and bad commits 
 # good: [98f7e32f20d28ec452afb208f9cffc08448a2652] Linux 6.11 
 git bisect good 98f7e32f20d28ec452afb208f9cffc08448a2652 
 # status: waiting for bad commit, 1 good commit known 
 # bad: [59b723cd2adbac2a34fc8e12c74ae26ae45bf230] Linux 6.12-rc6 
 git bisect bad 59b723cd2adbac2a34fc8e12c74ae26ae45bf230 
 # bad: [de848da12f752170c2ebe114804a985314fd5a6a] Merge tag 'drm-next- 
 2024-09-19' of https://gitlab.freedesktop.org/drm/kernel 
 git bisect bad de848da12f752170c2ebe114804a985314fd5a6a 
 # bad: [7b17f5ebd5fc5e9275eaa5af3d0771f2a7b01bbf] Merge tag 'soc-dt- 
 6.12' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc 
 git bisect bad 7b17f5ebd5fc5e9275eaa5af3d0771f2a7b01bbf 
 # good: [64dd3b6a79f0907d36de481b0f15fab323a53e5a] Merge tag 'for- 
 linus-non-x86' of git://git.kernel.org/pub/scm/virt/kvm/kvm 
 git bisect good 64dd3b6a79f0907d36de481b0f15fab323a53e5a 
 # bad: [daa394f0f9d3cb002c72e2d3db99972e2ee42862] Merge tag 'core- 
 debugobjects-2024-09-16' of 
 git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 
 git bisect bad daa394f0f9d3cb002c72e2d3db99972e2ee42862 
 # good: [effdcd5275ed645f6e0f8e8ce690b97795722197] Merge tag 'affs-for- 
 6.12-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux 
 git bisect good effdcd5275ed645f6e0f8e8ce690b97795722197 
 # bad: [26bb0d3f38a764b743a3ad5c8b6e5b5044d7ceb4] Merge tag 'for- 
 6.12/block-20240913' of git://git.kernel.dk/linux 
 git bisect bad 26bb0d3f38a764b743a3ad5c8b6e5b5044d7ceb4 
 # bad: [3a4d319a8fb5a9bbdf5b31ef32841eb286b1dcc2] Merge tag 'for- 
 6.12/io_uring-20240913' of git://git.kernel.dk/linux 
 git bisect bad 3a4d319a8fb5a9bbdf5b31ef32841eb286b1dcc2 
 # good: [df2825e98507d10cb037a308087ecd7cb3f6688d] btrfs: always pass 
 readahead state to defrag 
 git bisect good df2825e98507d10cb037a308087ecd7cb3f6688d 
 # good: [69a3a0a45a2f72412c2ba31761cc9193bb746fef] Merge tag 'erofs- 
 for-6.12-rc1' of 
 git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs 
 git bisect good 69a3a0a45a2f72412c2ba31761cc9193bb746fef 
 # good: [ecd5c9b29643f383d39320e30d21b8615bd893da] io_uring/kbuf: add 
 io_kbuf_commit() helper 
 git bisect good ecd5c9b29643f383d39320e30d21b8615bd893da 
 # good: [f011c9cf04c06f16b24f583d313d3c012e589e50] io_uring/sqpoll: do 
 not allow pinning outside of cpuset 
 git bisect good f011c9cf04c06f16b24f583d313d3c012e589e50 
 # bad: [84eacf177faa605853c58e5b1c0d9544b88c16fd] io_uring/io-wq: 
 inherit cpuset of cgroup in io worker 
 git bisect bad 84eacf177faa605853c58e5b1c0d9544b88c16fd 
 # bad: [90bfb28d5fa8127a113a140c9791ea0b40ab156a] io_uring/rw: drop - 
 EOPNOTSUPP check in __io_complete_rw_common() 
 git bisect bad 90bfb28d5fa8127a113a140c9791ea0b40ab156a 
 # good: [c0a9d496e0fece67db777bd48550376cf2960c47] io_uring/rw: treat - 
 EOPNOTSUPP for IOCB_NOWAIT like -EAGAIN 
 git bisect good c0a9d496e0fece67db777bd48550376cf2960c47 
 # first bad commit: [90bfb28d5fa8127a113a140c9791ea0b40ab156a] 
 io_uring/rw: drop -EOPNOTSUPP check in __io_complete_rw_common() 
  
 On Thu, 2025-10-02 at 16:25 +0200, Salvatore Bonaccorso wrote: 
 > Control: tags -1 + moreinfo 
 > 
 > Hi Kevin, 
 > 
 > On Thu, Oct 02, 2025 at 02:28:48PM +0300, Kevin Lumik wrote: 
 > > On Sat, 27 Sep 2025 16:09:08 +0200 Salvatore Bonaccorso 
 > >  wrote: 
 > > > - Can you temporarily test as well the upper kernels from 
 > > > unstable 
 > > > €€€€(6.16.8-1 right now, but soonish 6.16.9-1 as well) to see if 
 > > > the 
 > > > €€€€problem is triggerable there as well? 
 > > 
 > > I tested the upper kernels and was still able to reproduce the 
 > > issue on 
 > > both 6.16.8-1 and 6.16.9-1. 
 > 
 > Ok thanks for doing so, have just updated as well the metadata to 
 > reflect that. 
 > 
 > > > Note: If we find the problem is prosent in 6.16.y then we need to 
 > > > find 
 > > > the breaking change between 6.1.y and 6.12.y. It might be asier 
 > > > here 
 > > > if we first make the range of potential upstream versions more 
 > > > narrow, 
 > > > so you might first do a "manual bisect" on Debian revisions (For 
 > > > that 
 > > > use the snapshot.debian.org service to retrieve linux-images 
 > > > uploaded 
 > > > to Debian. E.g. go down from 6.12 in "major version" steps until 
 > > > we 
 > > > have close enough range of broken/fixed upstream major versions. 
 > > > Then 
 > > > perform a normal bisect between the upstream versions. 
 > > 
 > > Using snapshot.debian.org I have determined that the breaking 
 > > change was 
 > > introduced between 6.11.10 (tested 6.11.10-1) and 6.12 (tested 
 > > 6.12.3-1 
 > > and 6.12~rc6-1~exp1). 
 > > 
 > > Thank you for your assistance. Let me know what would be the next 
 > > step. 
 > 
 > Okay so we have a good range to look at. The next steps would be to 
 > bisect the changes. Actually it would be good to first test 6.11 
 > itself upstream. So the steps to do would be roughly the following, 
 > which needs compiling and testing a few kernels: 
 > 
 > €€€€€€ git clone 
 > https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git 
 > €€€€€€ cd linux-stable 
  
 [continued in next message] 
  
 --- SoupGate-Win32 v1.05 
  * Origin: you cannot sedate... all the things you hate (1:229/2)