Re: [PATCH v5 07/11] iomap: fix iomap_dio_zero() for fs bs > system page size

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, May 07, 2024 at 09:10:15AM -0700, Christoph Hellwig wrote:
> On Tue, May 07, 2024 at 05:00:28PM +0100, Matthew Wilcox wrote:
> > If the len is more than PAGE_SIZE * BIO_MAX_VECS, __bio_add_page()
> > will fail silently.  I hate this interface.
> 
> No, it won't.  You can pass an arbitray len to it.
> 
> > 
> > You should be doing something like ...
> > 
> > 	while (len) {
> > 		unsigned int io_len = min_t(unsigned int, len, PAGE_SIZE);
> > 
> > 		while (!bio || bio_add_page() < io_len) {
> > 			if (bio)
> > 				iomap_dio_submit_bio(iter, dio, bio, pos);
> > 			bio = iomap_dio_alloc_bio(iter, dio, BIO_MAX_VECS,
> > 					REQ_OP_WRITE | REQ_SYNC | REQ_IDLE);
> > 		 	fscrypt_set_bio_crypt_ctx(bio, inode,
> > 					pos >> inode->i_blkbits, GFP_KERNEL);
> > 		}
> > 	}
> 
> Wee, no.  The right way is:
> 
> 	bio = iomap_dio_alloc_bio(iter, dio, 1,
> 			REQ_OP_WRITE | REQ_SYNC | REQ_IDLE);
> 	__bio_add_page(bio, page, len, 0);

no?  len can be > PAGE_SIZE.  and that can be true in the folio version
too, because we won't necessarily be able to allocate the THP.

> 	fscrypt_set_bio_crypt_ctx(bio, inode,
> 			pos >> inode->i_blkbits, GFP_KERNEL);
> 
> (or even better the folio version)




[Index of Archives]     [Linux Ext4 Filesystem]     [Union Filesystem]     [Filesystem Testing]     [Ceph Users]     [Ecryptfs]     [NTFS 3]     [AutoFS]     [Kernel Newbies]     [Share Photos]     [Security]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux Cachefs]     [Reiser Filesystem]     [Linux RAID]     [NTFS 3]     [Samba]     [Device Mapper]     [CEPH Development]

  Powered by Linux