Re: [PATCH 2/2] block: make bio auto-integrity deadlock safe

[hch <hch@lst.de>]

On Mon, Nov 03, 2025 at 01:40:10PM +0000, Johannes Thumshirn wrote:
> On 11/3/25 11:18 AM, Christoph Hellwig wrote:
> > +void bio_integrity_alloc_buf(struct bio *bio, bool zero_buffer)
> > +{
> > +	struct blk_integrity *bi = blk_get_integrity(bio->bi_bdev->bd_disk);
> > +	struct bio_integrity_payload *bip = bio_integrity(bio);
> > +	unsigned int len = bio_integrity_bytes(bi, bio_sectors(bio));
> > +	gfp_t gfp = GFP_NOIO | (zero_buffer ? __GFP_ZERO : 0);
> > +	void *buf;
> > +
> > +	buf = kmalloc(len, (gfp & ~__GFP_DIRECT_RECLAIM) |
> > +			__GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN);
> 
> 
> Why can't we clear the flags when assigning gfp, or at least outside of 
> kmalloc()s parameter list?

We could, but what's the point?

> > +++ b/include/linux/bio-integrity.h
> > @@ -14,6 +14,8 @@ enum bip_flags {
> >   	BIP_CHECK_REFTAG	= 1 << 6, /* reftag check */
> >   	BIP_CHECK_APPTAG	= 1 << 7, /* apptag check */
> >   	BIP_P2P_DMA		= 1 << 8, /* using P2P address */
> > +
> > +	BIP_MEMPOOL		= 1 << 15, /* buffer backed by mempool */
> >   };
> >   
> 
> Any specific reason for the hole?

Because it's really just an internal flag.  (So is BIP_P2P_DMA, but
that should go away this merge window).