Re: [PATCH v2] blk-crypto-fallback: Fix deadlock when encrypting large bio in dm layer

On Thu, Aug 13, 2026 at 11:34:49PM -0700, Christoph Hellwig wrote:
> On Tue, Aug 11, 2026 at 03:03:18PM -0700, Eric Biggers wrote:
> > Encrypting a large bio with more than BIO_MAX_VECS pages can still
> > deadlock, even after it was attempted to be fixed by
> > commit b37fbce460ad ("blk-crypto: optimize bio splitting in
> > blk_crypto_fallback_encrypt_bio") and commit 3d939695e682 ("blk-crypto:
> > use mempool_alloc_bulk for encrypted bio page allocation").
> > 
> > This is because __blk_crypto_fallback_encrypt_bio() assumes that the
> > bounce bios that it submits will eventually complete, unblocking it from
> > allocating additional bounce bios and pages.  However, dm-inlinecrypt.c
> > calls __blk_crypto_submit_bio() from within submit_bio() itself.  In
> > this case, the recursive submit_bio() simply adds the bounce bio to
> > current->bio_list without actually submitting it yet.  That breaks the
> > guarantee that forward progress is being made.
> > 
> > To fix this, allocate the bio and bounce pages with GFP_NOWAIT if
> > current->bio_list is set.  If it fails, punt the encryption of the
> > remaining part of the bio to a kworker.
> > 
> > Fixes: 488f6682c832 ("block: blk-crypto-fallback for Inline Encryption")
> 
> No, it dos not fix that commit.  I "fixes" the misguided attempt to use
> this from inside stackable drivers.  Let's drop the dm-inlinecrypt
> target instead as this just shows up one more reason why that is a
> broken idea.
> 
> > Cc: [email protected]
> > Signed-off-by: Eric Biggers <[email protected]>
> > ---
> > 
> > Changed in v2:
> > - Switched to the less efficient approach preferred by Christoph
> 
> This is in no way the approach I prefer or even advocated for.

It is what you asked for on v1 and you've now changed your position, as
far as I can tell.  I can change the Fixes commit to e7f57d2c47e2.
dm-inlinecrypt is useful and it does not make sense to drop it.

- Eric