From: Chaitanya Kulkarni chaitanya.kulkarni@wdc.com
mainline inclusion from mainline-5.9-rc1 commit 61f3b89630973037f67d8e25e5d26e80a51a7b37 category: bugfix bugzilla: 175286 CVE: NA
---------------------------
The NVMe PCIe declares module parameter io_queue_depth as int. Change this to u16 as queue depth can never be negative. Now to reflect this update module parameter getter function from param_get_int() -> param_get_uint() and respective setter function with type of n changed from int to u16 with param_set_int() to param_set_ushort(). Finally update struct nvme_dev q_depth member to u16 and use u16 in min_t() when calculating dev->q_depth in the nvme_pci_enable() (since q_depth is now u16) and use unsigned int instead of int when calculating dev->tagset.queue_depth as target variable tagset->queue_depth is of type unsigned int in nvme_dev_add().
Signed-off-by: Chaitanya Kulkarni chaitanya.kulkarni@wdc.com Reviewed-by: Sagi Grimberg sagi@grimberg.me Signed-off-by: Christoph Hellwig hch@lst.de
Conflicts: drivers/nvme/host/pci.c [ Feature patch 3b6592f70ad7("nvme: utilize two queue maps, one for reads and one for writes") is not applied. ed92ad37e88("nvme-pci: only set nr_maps to 2 if poll queues are supported") is not applied. aa22c8e665("nvme-pci: set ctrl sqsize to the device q_depth") is not applied. ]
Signed-off-by: Zhihao Cheng chengzhihao1@huawei.com Reviewed-by: Hou Tao houtao1@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/nvme/host/pci.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 0a5e20808ba2e..e741b1d167c3e 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -66,10 +66,10 @@ MODULE_PARM_DESC(sgl_threshold, static int io_queue_depth_set(const char *val, const struct kernel_param *kp); static const struct kernel_param_ops io_queue_depth_ops = { .set = io_queue_depth_set, - .get = param_get_int, + .get = param_get_uint, };
-static int io_queue_depth = 1024; +static unsigned int io_queue_depth = 1024; module_param_cb(io_queue_depth, &io_queue_depth_ops, &io_queue_depth, 0644); MODULE_PARM_DESC(io_queue_depth, "set io queue depth, should >= 2");
@@ -93,7 +93,7 @@ struct nvme_dev { unsigned online_queues; unsigned max_qid; unsigned int num_vecs; - int q_depth; + u16 q_depth; u32 db_stride; void __iomem *bar; unsigned long bar_mapped_size; @@ -126,13 +126,14 @@ struct nvme_dev {
static int io_queue_depth_set(const char *val, const struct kernel_param *kp) { - int n = 0, ret; + int ret; + u16 n;
- ret = kstrtoint(val, 10, &n); + ret = kstrtou16(val, 10, &n); if (ret != 0 || n < 2) return -EINVAL;
- return param_set_int(val, kp); + return param_set_ushort(val, kp); }
static inline unsigned int sq_idx(unsigned int qid, u32 stride) @@ -2062,8 +2063,8 @@ static int nvme_dev_add(struct nvme_dev *dev) dev->tagset.nr_hw_queues = dev->online_queues - 1; dev->tagset.timeout = NVME_IO_TIMEOUT; dev->tagset.numa_node = dev_to_node(dev->dev); - dev->tagset.queue_depth = - min_t(int, dev->q_depth, BLK_MQ_MAX_DEPTH) - 1; + dev->tagset.queue_depth = min_t(unsigned int, dev->q_depth, + BLK_MQ_MAX_DEPTH) - 1; dev->tagset.cmd_size = nvme_pci_cmd_size(dev, false); if ((dev->ctrl.sgls & ((1 << 0) | (1 << 1))) && sgl_threshold) { dev->tagset.cmd_size = max(dev->tagset.cmd_size, @@ -2121,7 +2122,7 @@ static int nvme_pci_enable(struct nvme_dev *dev)
dev->ctrl.cap = lo_hi_readq(dev->bar + NVME_REG_CAP);
- dev->q_depth = min_t(int, NVME_CAP_MQES(dev->ctrl.cap) + 1, + dev->q_depth = min_t(u16, NVME_CAP_MQES(dev->ctrl.cap) + 1, io_queue_depth); dev->db_stride = 1 << NVME_CAP_STRIDE(dev->ctrl.cap); dev->dbs = dev->bar + 4096;
From: John Garry john.garry@huawei.com
mainline inclusion from mainline-5.9-rc3 commit 7442ddcedc344b6fa073692f165dffdd1889e780 category: bugfix bugzilla: 175286 CVE: NA
---------------------------
Recently nvme_dev.q_depth was changed from an int to u16 type.
This falls over for the queue depth calculation in nvme_pci_enable(), where NVME_CAP_MQES(dev->ctrl.cap) + 1 may overflow as a u16, as NVME_CAP_MQES() is a 16b number also. That happens for me, and this is the result:
root@ubuntu:/home/john# [148.272996] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000010 Mem abort info: ESR = 0x96000004 EC = 0x25: DABT (current EL), IL = 32 bits SET = 0, FnV = 0 EA = 0, S1PTW = 0 Data abort info: ISV = 0, ISS = 0x00000004 CM = 0, WnR = 0 user pgtable: 4k pages, 48-bit VAs, pgdp=00000a27bf3c9000 [0000000000000010] pgd=0000000000000000, p4d=0000000000000000 Internal error: Oops: 96000004 [#1] PREEMPT SMP Modules linked in: nvme nvme_core CPU: 56 PID: 256 Comm: kworker/u195:0 Not tainted 5.8.0-next-20200812 #27 Hardware name: Huawei D06 /D06, BIOS Hisilicon D06 UEFI RC0 - V1.16.01 03/15/2019 Workqueue: nvme-reset-wq nvme_reset_work [nvme] pstate: 80c00009 (Nzcv daif +PAN +UAO BTYPE=--) pc : __sg_alloc_table_from_pages+0xec/0x238 lr : __sg_alloc_table_from_pages+0xc8/0x238 sp : ffff800013ccbad0 x29: ffff800013ccbad0 x28: ffff0a27b3d380a8 x27: 0000000000000000 x26: 0000000000002dc2 x25: 0000000000000dc0 x24: 0000000000000000 x23: 0000000000000000 x22: ffff800013ccbbe8 x21: 0000000000000010 x20: 0000000000000000 x19: 00000000fffff000 x18: ffffffffffffffff x17: 00000000000000c0 x16: fffffe289eaf6380 x15: ffff800011b59948 x14: ffff002bc8fe98f8 x13: ff00000000000000 x12: ffff8000114ca000 x11: 0000000000000000 x10: ffffffffffffffff x9 : ffffffffffffffc0 x8 : ffff0a27b5f9b6a0 x7 : 0000000000000000 x6 : 0000000000000001 x5 : ffff0a27b5f9b680 x4 : 0000000000000000 x3 : ffff0a27b5f9b680 x2 : 0000000000000000 x1 : 0000000000000001 x0 : 0000000000000000 Call trace: __sg_alloc_table_from_pages+0xec/0x238 sg_alloc_table_from_pages+0x18/0x28 iommu_dma_alloc+0x474/0x678 dma_alloc_attrs+0xd8/0xf0 nvme_alloc_queue+0x114/0x160 [nvme] nvme_reset_work+0xb34/0x14b4 [nvme] process_one_work+0x1e8/0x360 worker_thread+0x44/0x478 kthread+0x150/0x158 ret_from_fork+0x10/0x34 Code: f94002c3 6b01017f 540007c2 11000486 (f8645aa5) ---[ end trace 89bb2b72d59bf925 ]---
Fix by making onto a u32.
Also use u32 for nvme_dev.q_depth, as we assign this value from nvme_dev.q_depth, and nvme_dev.q_depth will possibly hold 65536 - this avoids the same crash as above.
Fixes: 61f3b8963097 ("nvme-pci: use unsigned for io queue depth") Signed-off-by: John Garry john.garry@huawei.com Reviewed-by: Keith Busch kbusch@kernel.org Signed-off-by: Sagi Grimberg sagi@grimberg.me Signed-off-by: Jens Axboe axboe@kernel.dk
Conflicts: drivers/nvme/host/pci.c [ Feature patch 3b6592f70ad7("nvme: utilize two queue maps, one for reads and one for writes") is not applied. aa22c8e665("nvme-pci: set ctrl sqsize to the device q_depth") is not applied. ]
Signed-off-by: Zhihao Cheng chengzhihao1@huawei.com Reviewed-by: Hou Tao houtao1@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/nvme/host/pci.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index e741b1d167c3e..d1aab5fe651c6 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -93,7 +93,7 @@ struct nvme_dev { unsigned online_queues; unsigned max_qid; unsigned int num_vecs; - u16 q_depth; + u32 q_depth; u32 db_stride; void __iomem *bar; unsigned long bar_mapped_size; @@ -127,13 +127,13 @@ struct nvme_dev { static int io_queue_depth_set(const char *val, const struct kernel_param *kp) { int ret; - u16 n; + u32 n;
- ret = kstrtou16(val, 10, &n); + ret = kstrtou32(val, 10, &n); if (ret != 0 || n < 2) return -EINVAL;
- return param_set_ushort(val, kp); + return param_set_uint(val, kp); }
static inline unsigned int sq_idx(unsigned int qid, u32 stride) @@ -167,7 +167,7 @@ struct nvme_queue { dma_addr_t sq_dma_addr; dma_addr_t cq_dma_addr; u32 __iomem *q_db; - u16 q_depth; + u32 q_depth; u16 cq_vector; u16 sq_tail; u16 cq_head; @@ -2122,7 +2122,7 @@ static int nvme_pci_enable(struct nvme_dev *dev)
dev->ctrl.cap = lo_hi_readq(dev->bar + NVME_REG_CAP);
- dev->q_depth = min_t(u16, NVME_CAP_MQES(dev->ctrl.cap) + 1, + dev->q_depth = min_t(u32, NVME_CAP_MQES(dev->ctrl.cap) + 1, io_queue_depth); dev->db_stride = 1 << NVME_CAP_STRIDE(dev->ctrl.cap); dev->dbs = dev->bar + 4096;
-
Yang Yingliang