From 8098514bd5ca98beca6ec725751d82d0d5b492d8 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Fri, 9 May 2025 08:38:02 -0700 Subject: [PATCH] block: always allocate integrity buffer when required Many nvme metadata formats can not strip or generate the metadata on the controller side. For these formats, a host provided integrity buffer is mandatory even if it isn't checked. The block integrity read_verify and write_generate attributes prevent allocating the metadata buffer, but we need it when the format requires it, otherwise reads and writes will be rejected by the driver with IO errors. Assume the integrity buffer can be offloaded to the controller if the metadata size is the same as the protection information size. Otherwise provide an unchecked host buffer when the read verify or write generation attributes are disabled. This fixes the following nvme warning: ------------[ cut here ]------------ WARNING: CPU: 1 PID: 371 at drivers/nvme/host/core.c:1036 nvme_setup_rw+0x122/0x210 ... RIP: 0010:nvme_setup_rw+0x122/0x210 ... Call Trace: nvme_setup_cmd+0x1b4/0x280 nvme_queue_rqs+0xc4/0x1f0 [nvme] blk_mq_dispatch_queue_requests+0x24a/0x430 blk_mq_flush_plug_list+0x50/0x140 __blk_flush_plug+0xc1/0x100 __submit_bio+0x1c1/0x360 ? submit_bio_noacct_nocheck+0x2d6/0x3c0 submit_bio_noacct_nocheck+0x2d6/0x3c0 ? submit_bio_noacct+0x47/0x4c0 submit_bio_wait+0x48/0xa0 __blkdev_direct_IO_simple+0xee/0x210 ? current_time+0x1d/0x100 ? current_time+0x1d/0x100 ? __bio_clone+0xb0/0xb0 blkdev_read_iter+0xbb/0x140 vfs_read+0x239/0x310 ksys_read+0x58/0xc0 do_syscall_64+0x6c/0x180 entry_SYSCALL_64_after_hwframe+0x4b/0x53 Signed-off-by: Keith Busch Reviewed-by: Martin K. Petersen Reviewed-by: Christoph Hellwig Link: https://lore.kernel.org/r/20250509153802.3482493-1-kbusch@meta.com Signed-off-by: Jens Axboe --- block/bio-integrity-auto.c | 62 +++++++++++++++++++++++++++++--------- 1 file changed, 47 insertions(+), 15 deletions(-) diff --git a/block/bio-integrity-auto.c b/block/bio-integrity-auto.c index e524c609be506..9c66576647920 100644 --- a/block/bio-integrity-auto.c +++ b/block/bio-integrity-auto.c @@ -9,6 +9,7 @@ * not aware of PI. */ #include +#include #include #include "blk.h" @@ -43,6 +44,29 @@ static void bio_integrity_verify_fn(struct work_struct *work) bio_endio(bio); } +#define BIP_CHECK_FLAGS (BIP_CHECK_GUARD | BIP_CHECK_REFTAG | BIP_CHECK_APPTAG) +static bool bip_should_check(struct bio_integrity_payload *bip) +{ + return bip->bip_flags & BIP_CHECK_FLAGS; +} + +static bool bi_offload_capable(struct blk_integrity *bi) +{ + switch (bi->csum_type) { + case BLK_INTEGRITY_CSUM_CRC64: + return bi->tuple_size == sizeof(struct crc64_pi_tuple); + case BLK_INTEGRITY_CSUM_CRC: + case BLK_INTEGRITY_CSUM_IP: + return bi->tuple_size == sizeof(struct t10_pi_tuple); + default: + pr_warn_once("%s: unknown integrity checksum type:%d\n", + __func__, bi->csum_type); + fallthrough; + case BLK_INTEGRITY_CSUM_NONE: + return false; + } +} + /** * __bio_integrity_endio - Integrity I/O completion function * @bio: Protected bio @@ -54,12 +78,12 @@ static void bio_integrity_verify_fn(struct work_struct *work) */ bool __bio_integrity_endio(struct bio *bio) { - struct blk_integrity *bi = blk_get_integrity(bio->bi_bdev->bd_disk); struct bio_integrity_payload *bip = bio_integrity(bio); struct bio_integrity_data *bid = container_of(bip, struct bio_integrity_data, bip); - if (bio_op(bio) == REQ_OP_READ && !bio->bi_status && bi->csum_type) { + if (bio_op(bio) == REQ_OP_READ && !bio->bi_status && + bip_should_check(bip)) { INIT_WORK(&bid->work, bio_integrity_verify_fn); queue_work(kintegrityd_wq, &bid->work); return false; @@ -84,6 +108,7 @@ bool bio_integrity_prep(struct bio *bio) { struct blk_integrity *bi = blk_get_integrity(bio->bi_bdev->bd_disk); struct bio_integrity_data *bid; + bool set_flags = true; gfp_t gfp = GFP_NOIO; unsigned int len; void *buf; @@ -100,19 +125,24 @@ bool bio_integrity_prep(struct bio *bio) switch (bio_op(bio)) { case REQ_OP_READ: - if (bi->flags & BLK_INTEGRITY_NOVERIFY) - return true; + if (bi->flags & BLK_INTEGRITY_NOVERIFY) { + if (bi_offload_capable(bi)) + return true; + set_flags = false; + } break; case REQ_OP_WRITE: - if (bi->flags & BLK_INTEGRITY_NOGENERATE) - return true; - /* * Zero the memory allocated to not leak uninitialized kernel * memory to disk for non-integrity metadata where nothing else * initializes the memory. */ - if (bi->csum_type == BLK_INTEGRITY_CSUM_NONE) + if (bi->flags & BLK_INTEGRITY_NOGENERATE) { + if (bi_offload_capable(bi)) + return true; + set_flags = false; + gfp |= __GFP_ZERO; + } else if (bi->csum_type == BLK_INTEGRITY_CSUM_NONE) gfp |= __GFP_ZERO; break; default: @@ -137,19 +167,21 @@ bool bio_integrity_prep(struct bio *bio) bid->bip.bip_flags |= BIP_BLOCK_INTEGRITY; bip_set_seed(&bid->bip, bio->bi_iter.bi_sector); - if (bi->csum_type == BLK_INTEGRITY_CSUM_IP) - bid->bip.bip_flags |= BIP_IP_CHECKSUM; - if (bi->csum_type) - bid->bip.bip_flags |= BIP_CHECK_GUARD; - if (bi->flags & BLK_INTEGRITY_REF_TAG) - bid->bip.bip_flags |= BIP_CHECK_REFTAG; + if (set_flags) { + if (bi->csum_type == BLK_INTEGRITY_CSUM_IP) + bid->bip.bip_flags |= BIP_IP_CHECKSUM; + if (bi->csum_type) + bid->bip.bip_flags |= BIP_CHECK_GUARD; + if (bi->flags & BLK_INTEGRITY_REF_TAG) + bid->bip.bip_flags |= BIP_CHECK_REFTAG; + } if (bio_integrity_add_page(bio, virt_to_page(buf), len, offset_in_page(buf)) < len) goto err_end_io; /* Auto-generate integrity metadata if this is a write */ - if (bio_data_dir(bio) == WRITE) + if (bio_data_dir(bio) == WRITE && bip_should_check(&bid->bip)) blk_integrity_generate(bio); else bid->saved_bio_iter = bio->bi_iter; -- 2.39.5