linux-dev/drivers/misc/vmw_vmci, branch master

misc/vmw_vmci: Use kmap_local_page() in vmci_queue_pair.c

2022-09-09T08:22:32Z

kmap() is being deprecated in favor of kmap_local_page(). There are two main problems with kmap(): (1) It comes with an overhead as the mapping space is restricted and protected by a global lock for synchronization and (2) it also requires global TLB invalidation when the kmap’s pool wraps and it might block when the mapping space is fully utilized until a slot becomes available. With kmap_local_page() the mappings are per thread, CPU local, can take page faults, and can be called from any context (including interrupts). It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore, the tasks can be preempted and, when they are scheduled to run again, the kernel virtual addresses are restored and still valid. Since its use in vmci_queue_pair.c is safe everywhere, replace kmap() with kmap_local_page(). Cc: "Venkataramanan, Anirudh" Suggested-by: Ira Weiny Signed-off-by: Fabio M. De Francesco Link: https://lore.kernel.org/r/20220901135714.16481-1-fmdefrancesco@gmail.com Signed-off-by: Greg Kroah-Hartman

VMCI: Add support for ARM64

2022-04-24T15:32:14Z

Add support for ARM64 architecture so that the driver can now be built and VMCI device can be used. Update Kconfig file to allow the driver to be built on ARM64 as well. Fail vmci_guest_probe_device() on ARM64 if the device does not support MMIO register access. Lastly, add virtualization specific barriers which map to actual memory barrier instructions on ARM64, because it is required in case of ARM64 for queuepair (de)queuing. Reviewed-by: Bryan Tan Reviewed-by: Cyprien Laplace Signed-off-by: Vishnu Dasa Link: https://lore.kernel.org/r/20220414193316.14356-1-vdasa@vmware.com Signed-off-by: Greg Kroah-Hartman

misc: vmw_vmci: replace usage of found with dedicated list iterator variable

2022-04-24T15:24:09Z

To move the list iterator variable into the list_for_each_entry_*() macro in the future it should be avoided to use the list iterator variable after the loop body. To *never* use the list iterator variable after the loop it was concluded to use a separate iterator variable instead of a found boolean [1]. This removes the need to use a found variable and simply checking if the variable was set, can determine if the break/goto was hit. Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ Signed-off-by: Jakob Koschel Link: https://lore.kernel.org/r/20220324073151.66305-1-jakobkoschel@gmail.com Signed-off-by: Greg Kroah-Hartman

VMCI: Release notification_bitmap in error path

2022-03-18T12:47:48Z

notification_bitmap may not be released when VMCI_CAPS_DMA_DATAGRAM capability is missing from the device. Add missing 'err_free_notification_bitmap' label and use it instead of 'err_free_data_buffers' to avoid this. Fixes: eed2298d9360 ("VMCI: dma dg: detect DMA datagram capability") Reported-by: Dan Carpenter Reviewed-by: Bryan Tan Reviewed-by: Rajesh Jalisatgi Signed-off-by: Vishnu Dasa Link: https://lore.kernel.org/r/20220318060040.31621-1-vdasa@vmware.com Signed-off-by: Greg Kroah-Hartman

VMCI: Check exclusive_vectors when freeing interrupt 1

2022-03-18T12:47:34Z

free_irq() may be called to free an interrupt that was not allocated. Add missing 'if' statement to check for exclusive_vectors when freeing interrupt 1. Fixes: cc68f2177fcb ("VMCI: dma dg: register dummy IRQ handlers for DMA datagrams") Reported-by: Dan Carpenter Reviewed-by: Bryan Tan Reviewed-by: Rajesh Jalisatgi Signed-off-by: Vishnu Dasa Link: https://lore.kernel.org/r/20220318055843.30606-1-vdasa@vmware.com Signed-off-by: Greg Kroah-Hartman

VMCI: Fix some error handling paths in vmci_guest_probe_device()

2022-02-25T10:55:01Z

The 'err_remove_vmci_dev_g' error label is not at the right place. This could lead to un-released resource. There is also a missing label. If pci_alloc_irq_vectors() fails, the previous vmci_event_subscribe() call must be undone. Acked-by: Vishnu Dasa Signed-off-by: Christophe JAILLET Link: https://lore.kernel.org/r/959218ce3b135197946d85cd9453551cd04fa5da.1645734041.git.christophe.jaillet@wanadoo.fr Signed-off-by: Greg Kroah-Hartman

VMCI: No need to clear memory after a dma_alloc_coherent() call

2022-02-25T10:55:01Z

dma_alloc_coherent() already clear the allocated memory, there is no need to explicitly call memset(). This saves a few cycles and a few lines of code. Acked-by: Vishnu Dasa Signed-off-by: Christophe JAILLET Link: https://lore.kernel.org/r/3e8c0bfaa77500e22d6c90f249da1c53dbae716e.1645734041.git.christophe.jaillet@wanadoo.fr Signed-off-by: Greg Kroah-Hartman

VMCI: Fix the description of vmci_check_host_caps()

2022-02-25T10:55:01Z

vmci_check_host_caps() doesn't return a bool but an int. Fix the description accordingly. Fixes: 782f24453536 ("VMCI: fix error handling path when registering guest driver") Acked-by: Vishnu Dasa Signed-off-by: Christophe JAILLET Link: https://lore.kernel.org/r/c181bec88aab1145d3868d61b7e52d53923f8206.1645734041.git.christophe.jaillet@wanadoo.fr Signed-off-by: Greg Kroah-Hartman

VMCI: dma dg: add support for DMA datagrams receive

2022-02-08T11:15:59Z

Use the DMA based receive operation instead of the ioread8_rep based datagram receive when DMA datagrams are supported. In the receive operation, configure the header to point to the page aligned VMCI_MAX_DG_SIZE part of the receive buffer using s/g configuration for the header. This ensures that the existing dispatch routine can be used with little modification. Initiate the receive by writing the lower 32 bit of the buffer to the VMCI_DATA_IN_LOW_ADDR register, and wait for the busy flag to be changed by the device using a wait queue. The existing dispatch routine for received datagrams is reused for the DMA datagrams with a few modifications: - the receive buffer is always the maximum size for DMA datagrams (IO ports would try with a shorter buffer first to reduce overhead of the ioread8_rep operation). - for DMA datagrams, datagrams are provided contiguous in the buffer as opposed to IO port datagrams, where they can start on any page boundary Reviewed-by: Vishnu Dasa Signed-off-by: Jorgen Hansen Link: https://lore.kernel.org/r/20220207102725.2742-9-jhansen@vmware.com Signed-off-by: Greg Kroah-Hartman

VMCI: dma dg: add support for DMA datagrams sends

2022-02-08T11:15:59Z

Use DMA based send operation from the transmit buffer instead of the iowrite8_rep based datagram send when DMA datagrams are supported. The outgoing datagram is sent as inline data in the VMCI transmit buffer. Once the header has been configured, the send is initiated by writing the lower 32 bit of the buffer base address to the VMCI_DATA_OUT_LOW_ADDR register. Only then will the device process the header and the datagram itself. Following that, the driver busy waits (it isn't possible to sleep on the send path) for the header busy flag to change - indicating that the send is complete. Reviewed-by: Vishnu Dasa Signed-off-by: Jorgen Hansen Link: https://lore.kernel.org/r/20220207102725.2742-8-jhansen@vmware.com Signed-off-by: Greg Kroah-Hartman