diff options
| author | Shay Drory <shayd@nvidia.com> | 2021-06-03 16:19:39 +0300 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2021-06-16 11:59:44 +0200 |
| commit | fc57713afaca68efb1692f0156d7def53f149914 (patch) | |
| tree | 5a10e26a1d9674775bf03e9bba7524273b882447 /drivers/infiniband | |
| parent | 64f1fb6acc2ab95982fc4334f351d7576c26f313 (diff) | |
RDMA/mlx4: Do not map the core_clock page to user space unless enabled
commit 404e5a12691fe797486475fe28cc0b80cb8bef2c upstream.
Currently when mlx4 maps the hca_core_clock page to the user space there
are read-modifiable registers, one of which is semaphore, on this page as
well as the clock counter. If user reads the wrong offset, it can modify
the semaphore and hang the device.
Do not map the hca_core_clock page to the user space unless the device has
been put in a backwards compatibility mode to support this feature.
After this patch, mlx4 core_clock won't be mapped to user space on the
majority of existing devices and the uverbs device time feature in
ibv_query_rt_values_ex() will be disabled.
Fixes: 52033cfb5aab ("IB/mlx4: Add mmap call to map the hardware clock")
Link: https://lore.kernel.org/r/9632304e0d6790af84b3b706d8c18732bc0d5e27.1622726305.git.leonro@nvidia.com
Signed-off-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'drivers/infiniband')
| -rw-r--r-- | drivers/infiniband/hw/mlx4/main.c | 5 |
1 files changed, 1 insertions, 4 deletions
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index b8274c6fc43e..48864269f042 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -577,12 +577,9 @@ static int mlx4_ib_query_device(struct ib_device *ibdev, props->cq_caps.max_cq_moderation_count = MLX4_MAX_CQ_COUNT; props->cq_caps.max_cq_moderation_period = MLX4_MAX_CQ_PERIOD; - if (!mlx4_is_slave(dev->dev)) - err = mlx4_get_internal_clock_params(dev->dev, &clock_params); - if (uhw->outlen >= resp.response_length + sizeof(resp.hca_core_clock_offset)) { resp.response_length += sizeof(resp.hca_core_clock_offset); - if (!err && !mlx4_is_slave(dev->dev)) { + if (!mlx4_get_internal_clock_params(dev->dev, &clock_params)) { resp.comp_mask |= MLX4_IB_QUERY_DEV_RESP_MASK_CORE_CLOCK_OFFSET; resp.hca_core_clock_offset = clock_params.offset % PAGE_SIZE; } |