Re: [PATCH 1/1] Revert "iomap: fall back to buffered writes for invalidation failures"

[Matthew Wilcox]

On Thu, Feb 10, 2022 at 10:15:52AM +0000, Lee Jones wrote:
> On Wed, 09 Feb 2022, Darrick J. Wong wrote:
> 
> > On Wed, Feb 09, 2022 at 08:52:43AM +0000, Lee Jones wrote:
> > > This reverts commit 60263d5889e6dc5987dc51b801be4955ff2e4aa7.
> > > 
> > > Reverting since this commit opens a potential avenue for abuse.
> > 
> > What kind of abuse?  Did you conclude there's an avenue solely because
> > some combination of userspace rigging produced a BUG warning?  Or is
> > this a real problem that someone found?
> 
> Genuine question: Is the ability for userspace to crash the kernel
> not enough to cause concern?  I would have thought that we'd want to
> prevent this.

The kernel doesn't crash.  It's a BUG().  That means it kills the
task which caused the BUG().  If you've specified that the kernel should
crash on seeing a BUG(), well, you made that decision, and you get to
live with the consequences.

> The link provided doesn't contain any further analysis.  Only the
> reproducer and kernel configuration used, which are both too large to
> enter into a Git commit.

But not too large to put in an email.  Which you should have sent to
begin with, not a stupid reversion commit.

> > OH WAIT, you're running this on the Android 5.10 kernel, aren't you?
> > The BUG report came from page_buffers failing to find any buffer heads
> > attached to the page.
> > https://android.googlesource.com/kernel/common/+/refs/heads/android12-5.10-2022-02/fs/ext4/inode.c#2647
> 
> Yes, the H/W I have to prototype these on is a phone and the report
> that came in was specifically built against the aforementioned
> kernel.
> 
> > Yeah, don't care.
> 
> "There is nothing to worry about, as it's intended behaviour"?

No.  You've come in like a fucking meteorite full of arrogance and
ignorance.  Nobody's reacting well to you right now.  Start again,
write a good bug report in a new thread.