linux-dev/drivers/infiniband/hw/efa/efa_com.c, branch linus/master

RDMA/efa: CQ notifications

2021-10-06T22:47:18Z

This patch adds support for CQ notifications through the standard verbs api. In order to achieve that, a new event queue (EQ) object is introduced, which is in charge of reporting completion events to the driver. On driver load, EQs are allocated and their affinity is set to a single cpu. When a user app creates a CQ with a completion channel, the completion vector number is converted to a EQ number, which is in charge of reporting the CQ events. In addition, the CQ creation admin command now returns an offset for the CQ doorbell, which is mapped to the userspace provider and is used to arm the CQ when requested by the user. The EQs use a single doorbell (located on the registers BAR), which encodes the EQ number and arm as part of the doorbell value. The EQs are polled by the driver on each new EQE, and arm it when the poll is completed. Link: https://lore.kernel.org/r/20211003105605.29222-1-galpress@amazon.com Reviewed-by: Firas JahJah Reviewed-by: Yossi Leybovich Signed-off-by: Gal Pressman Signed-off-by: Jason Gunthorpe

RDMA/efa: Remove duplication of upper/lower_32_bits

2021-01-28T19:56:13Z

The EFA_DMA_ADDR_TO_UINT32_HIGH/LOW macros are the same as the kernel's upper/lower_32_bits, remove them and use kernel macros instead. Link: https://lore.kernel.org/r/20210126120702.9807-3-galpress@amazon.com Reviewed-by: Firas JahJah Signed-off-by: Gal Pressman Signed-off-by: Jason Gunthorpe

RDMA/efa: Remove redundant NULL pointer check of CQE

2021-01-28T19:56:12Z

A pointer to store the command completion must be provided as it is always used in efa_com_put_comp_ctx() to return the completion context back to the pool. Remove the NULL pointer check and the redundant 'status' field stored on the context as it could be retrieved from the completion itself. Link: https://lore.kernel.org/r/20210126120702.9807-2-galpress@amazon.com Reviewed-by: Firas JahJah Reviewed-by: Yossi Leybovich Signed-off-by: Gal Pressman Signed-off-by: Jason Gunthorpe

RDMA/hw/efa/efa_com: Stop using param description notation for non-params

2021-01-22T18:37:29Z

Fixes the following W=1 kernel build warning(s): drivers/infiniband/hw/efa/efa_com.c:801: warning: Excess function parameter 'note' description in 'efa_com_admin_q_comp_intr_handler' Link: https://lore.kernel.org/r/20210121094519.2044049-4-lee.jones@linaro.org Cc: Gal Pressman Cc: Yossi Leybovich Cc: Doug Ledford Cc: Jason Gunthorpe Cc: linux-rdma@vger.kernel.org Signed-off-by: Lee Jones Acked-by: Gal Pressman Signed-off-by: Jason Gunthorpe

RDMA/efa: Count admin commands errors

2020-05-02T23:32:14Z

Add a new stat that counts admin commands failures, which might help when debugging different issues. Link: https://lore.kernel.org/r/20200420062213.44577-4-galpress@amazon.com Reviewed-by: Daniel Kranzdorf Reviewed-by: Yossi Leybovich Signed-off-by: Gal Pressman Signed-off-by: Jason Gunthorpe

RDMA/efa: Properly document the interrupt mask register

2020-02-28T16:12:12Z

The fact that the LSB in the register is the enable bit should not be an implicit assumption between the driver and the device, properly document that in the register definition. Link: https://lore.kernel.org/r/20200225114010.21790-3-galpress@amazon.com Reviewed-by: Firas JahJah Reviewed-by: Yossi Leybovich Signed-off-by: Gal Pressman Signed-off-by: Jason Gunthorpe

RDMA/efa: Unified getters/setters for device structs bitmask access

2020-02-28T16:12:04Z

Use unified macros for device structs access instead of open coding the shifts and masks over and over again. Link: https://lore.kernel.org/r/20200225114010.21790-2-galpress@amazon.com Reviewed-by: Firas JahJah Reviewed-by: Yossi Leybovich Signed-off-by: Gal Pressman Signed-off-by: Jason Gunthorpe

RDMA/efa: Clear the admin command buffer prior to its submission

2019-11-14T15:57:33Z

We cannot rely on the entry memcpy as we only copy the actual size of the command, the rest of the bytes must be memset to zero. Currently providing non-zero memory will not have any user visible impact. However, since admin commands are extendable (in a backwards compatible way) everything beyond the size of the command must be cleared to prevent issues in the future. Fixes: 0420e542569b ("RDMA/efa: Implement functions that submit and complete admin commands") Link: https://lore.kernel.org/r/20191112092608.46964-1-galpress@amazon.com Reviewed-by: Daniel Kranzdorf Reviewed-by: Firas JahJah Signed-off-by: Gal Pressman Signed-off-by: Jason Gunthorpe

RDMA/efa: Rate limit admin queue error prints

2019-08-05T16:19:55Z

Admin queue error prints should never happen unless something wrong happened to the device. However, in the unfortunate case that it does, we should take extra care not to flood the log with error messages. Reviewed-by: Firas JahJah Reviewed-by: Yossi Leybovich Signed-off-by: Gal Pressman Link: https://lore.kernel.org/r/20190801171447.54440-3-galpress@amazon.com Signed-off-by: Doug Ledford

RDMA/efa: Entropy in admin commands id

2019-07-04T17:31:09Z

Make admin commands id easier to distinguish by using relevant bits from the producer counter. This allows us to differentiate admin commands with the same producer index (happens after admin queue overlap), which is helpful when debugging. Signed-off-by: Daniel Kranzdorf Reviewed-by: Firas JahJah Reviewed-by: Yossi Leybovich Signed-off-by: Gal Pressman Signed-off-by: Jason Gunthorpe