[ Upstream commit 08ae4b20f5e82101d77326ecab9089e110f224cc ]
The handling of the `COMEDI_INSNLIST` ioctl allocates a kernel buffer to hold the array of `struct comedi_insn`, getting the length from the `n_insns` member of the `struct comedi_insnlist` supplied by the user. The allocation will fail with a WARNING and a stack dump if it is too large.
Avoid that by failing with an `-EINVAL` error if the supplied `n_insns` value is unreasonable.
Define the limit on the `n_insns` value in the `MAX_INSNS` macro. Set this to the same value as `MAX_SAMPLES` (65536), which is the maximum allowed sum of the values of the member `n` in the array of `struct comedi_insn`, and sensible comedi instructions will have an `n` of at least 1.
Reported-by: syzbot+d6995b62e5ac7d79557a@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=d6995b62e5ac7d79557a Fixes: ed9eccbe8970 ("Staging: add comedi core") Tested-by: Ian Abbott abbotti@mev.co.uk Cc: stable@vger.kernel.org # 5.13+ Signed-off-by: Ian Abbott abbotti@mev.co.uk Link: https://lore.kernel.org/r/20250704120405.83028-1-abbotti@mev.co.uk Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org [ Reworked for before commit bac42fb21259 ("comedi: get rid of compat_alloc_user_space() mess in COMEDI_CMD{,TEST} compat") ] Signed-off-by: Ian Abbott abbotti@mev.co.uk --- drivers/staging/comedi/comedi_compat32.c | 3 +++ drivers/staging/comedi/comedi_fops.c | 13 +++++++++++++ 2 files changed, 16 insertions(+)
diff --git a/drivers/staging/comedi/comedi_compat32.c b/drivers/staging/comedi/comedi_compat32.c index 36a3564ba1fb..2f444e2b92c2 100644 --- a/drivers/staging/comedi/comedi_compat32.c +++ b/drivers/staging/comedi/comedi_compat32.c @@ -360,6 +360,9 @@ static int compat_insnlist(struct file *file, unsigned long arg) if (err) return -EFAULT;
+ if (n_insns > 65536) /* See MAX_INSNS in comedi_fops.c */ + return -EINVAL; + /* Allocate user memory to copy insnlist and insns into. */ s = compat_alloc_user_space(offsetof(struct combined_insnlist, insn[n_insns])); diff --git a/drivers/staging/comedi/comedi_fops.c b/drivers/staging/comedi/comedi_fops.c index 8b2337f8303d..413863bc929b 100644 --- a/drivers/staging/comedi/comedi_fops.c +++ b/drivers/staging/comedi/comedi_fops.c @@ -1502,6 +1502,16 @@ static int parse_insn(struct comedi_device *dev, struct comedi_insn *insn, return ret; }
+#define MAX_INSNS 65536 +static int check_insnlist_len(struct comedi_device *dev, unsigned int n_insns) +{ + if (n_insns > MAX_INSNS) { + dev_dbg(dev->class_dev, "insnlist length too large\n"); + return -EINVAL; + } + return 0; +} + /* * COMEDI_INSNLIST ioctl * synchronous instruction list @@ -1534,6 +1544,9 @@ static int do_insnlist_ioctl(struct comedi_device *dev, if (copy_from_user(&insnlist, arg, sizeof(insnlist))) return -EFAULT;
+ ret = check_insnlist_len(dev, insnlist32.n_insns); + if (ret) + return ret; insns = kcalloc(insnlist.n_insns, sizeof(*insns), GFP_KERNEL); if (!insns) { ret = -ENOMEM;
[ Sasha's backport helper bot ]
Hi,
Summary of potential issues: ❌ Build failures detected
The upstream commit SHA1 provided is correct: 08ae4b20f5e82101d77326ecab9089e110f224cc
Status in newer kernel trees: 6.15.y | Not found 6.12.y | Not found 6.6.y | Not found 6.1.y | Not found 5.15.y | Not found 5.10.y | Not found
Note: The patch differs from the upstream commit: --- 1: 08ae4b20f5e8 < -: ------------ comedi: Fail COMEDI_INSNLIST ioctl if n_insns is too large -: ------------ > 1: 480e150b6535 comedi: Fail COMEDI_INSNLIST ioctl if n_insns is too large
---
Results of testing on various branches:
| Branch | Patch Apply | Build Test | |---------------------------|-------------|------------| | origin/linux-5.4.y | Success | Failed |
Build Errors: origin/linux-5.4.y: Build error: Building current HEAD with log output Build x86: exited with code 2 Cleaning up worktrees... Cleaning up worktrees... Cleaning up worktrees... Cleaning up worktrees...
linux-stable-mirror@lists.linaro.org