linux/drivers/misc, branch v5.11

misc: rtsx: init value of aspm_enabled

2021-01-22T10:04:53Z

make sure ASPM state sync with pcr->aspm_enabled init value pcr->aspm_enabled Cc: stable@vger.kernel.org Signed-off-by: Ricky Wu Link: https://lore.kernel.org/r/20210122081906.19100-1-ricky_wu@realtek.com Fixes: d928061c3143 ("misc: rtsx: modify en/disable aspm function") Signed-off-by: Greg Kroah-Hartman

habanalabs: disable FW events on device removal

2021-01-21T18:30:22Z

When device is removed, we need to make sure the F/W won't send us any more events because during the remove process we disable the interrupts. Signed-off-by: Oded Gabbay

habanalabs: fix backward compatibility of idle check

2021-01-21T18:30:22Z

Need to take the lower 32 bits of the driver's 64-bit idle mask and put it in the legacy 32-bit variable that the userspace reads to know the idle mask. Signed-off-by: Oded Gabbay

habanalabs: zero pci counters packet before submit to FW

2021-01-21T18:30:22Z

Driver does not zero some pci counters packets before sending to FW. This causes an out of sync PI/CI between driver and FW. Signed-off-by: Ofir Bitton Reviewed-by: Oded Gabbay Signed-off-by: Oded Gabbay

habanalabs: prevent soft lockup during unmap

2021-01-12T13:00:10Z

When using Deep learning framework such as tensorflow or pytorch, there are tens of thousands of host memory mappings. When the user frees all those mappings at the same time, the process of unmapping and unpinning them can take a long time, which may cause a soft lockup bug. To prevent this, we need to free the core to do other things during the unmapping process. For now, we chose to do it every 32K unmappings (each unmap is a single 4K page). Signed-off-by: Oded Gabbay

habanalabs: fix reset process in case of failures

2021-01-12T12:59:52Z

There are some points in the reset process where if the code fails for some reason, and the system admin tries to initiate the reset process again we will get a kernel panic. This is because there aren't any protections in different fini functions that are called during the reset process. The protections that are added in this patch make sure that if the fini functions are called multiple times, without calling init functions between them, there won't be double release of already released resources. Signed-off-by: Oded Gabbay

habanalabs: fix dma_addr passed to dma_mmap_coherent

2021-01-12T12:59:36Z

When doing dma_alloc_coherent in the driver, we add a certain hard-coded offset to the DMA address before returning to the callee function. This offset is needed when our device use this DMA address to perform outbound transactions to the host. However, if we want to map the DMA'able memory to the user via dma_mmap_coherent(), we need to pass the original dma address, without this offset. Otherwise, we will get erronouos mapping. Signed-off-by: Oded Gabbay

misc: pvpanic: Check devm_ioport_map() for NULL

2021-01-07T19:28:01Z

Inconveniently devm_ioport_map() and devm_ioremap_resource() return errors differently, i.e. former uses simply NULL pointer, while the latter an error pointer. Due to this, we have to check each of them separately. Fixes: f104060813fe ("misc: pvpanic: Combine ACPI and platform drivers") Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20201228184313.57610-1-andriy.shevchenko@linux.intel.com Signed-off-by: Greg Kroah-Hartman

habanalabs: Fix memleak in hl_device_reset

2020-12-29T21:23:12Z

When kzalloc() fails, we should execute hl_mmu_fini() to release the MMU module. It's the same when hl_ctx_init() fails. Signed-off-by: Dinghao Liu Reviewed-by: Oded Gabbay Signed-off-by: Oded Gabbay

habanalabs: fix order of status check

2020-12-28T06:47:39Z

When the device is in reset or needs to be reset, the disabled property is don't-care. Signed-off-by: Oded Gabbay