linux-dev/drivers/s390/cio/qdio_setup.c, branch linus/master

s390/qdio: clarify handler logic for qdio_handle_activate_check()

2021-12-06T13:42:25Z

qdio_handle_activate_check() tries to re-use one of the queue-specific handlers to report that the ACTIVATE ccw has been terminated. But the logic to select that handler is overly complex - in practice both qdio drivers have at least one Input Queue, so we never take the other paths. Make things more obvious by removing this unused code, and clearly spelling out that we re-use the Input Handler for generic error reporting. This also paves the way for a world without queue-specific error handlers. Signed-off-by: Julian Wiedmann Reviewed-by: Benjamin Block Signed-off-by: Heiko Carstens

s390/qdio: avoid allocating the qdio_irq with GFP_DMA

2021-12-06T13:42:24Z

The qdio_irq contains only two fields that are directly exposed to the HW (ccw and qib). And only the ccw needs to reside in 31-bit memory. So allocate it separately, and remove the GFP_DMA constraint from the qdio_irq allocation. Signed-off-by: Julian Wiedmann Reviewed-by: Benjamin Block Signed-off-by: Heiko Carstens

s390/qdio: improve handling of CIWs

2021-12-06T13:42:24Z

Fetch the individual CIWs when we actually need them, rather than fetching both of them in qdio_setup_irq() and then needing to cache them inside the qdio_irq. Also deal with the error when a CIW is not available, instead of silently dropping this error condition in qdio_setup_irq()'s caller. Signed-off-by: Julian Wiedmann Reviewed-by: Benjamin Block Signed-off-by: Heiko Carstens

s390/qeth: move qdio's QAOB cache into qeth

2021-10-25T12:58:19Z

qdio.ko no longer needs to care about how the QAOBs are allocated, from its perspective they are merely another parameter to do_QDIO(). So for a start, shift the cache into the only qdio driver that uses QAOBs (ie. qeth). Here there's further opportunity to optimize its usage in the future - eg. make it per-{device, TX queue}, or only compile it when the driver is built with CQ/QAOB support. Signed-off-by: Julian Wiedmann Reviewed-by: Benjamin Block Signed-off-by: David S. Miller

s390/qdio: remove unused support for SLIB parameters

2021-08-18T08:01:28Z

Neither of the two drivers provides any SLIB parameter data, so get rid of the dead code. Signed-off-by: Julian Wiedmann Reviewed-by: Benjamin Block Signed-off-by: Vasily Gorbik

s390/qdio: consolidate QIB code

2021-08-18T08:01:28Z

Move all QIB-related code into qdio_setup_qib(), and slightly re-order it according to the order of the struct's fields. This makes it easier to understand what the QIB actually looks like before we send it to HW. Also get rid of the qebsm_possible() helper - as 31-bit support is long gone, the comment doesn't make any sense. And while removing some stale QIB-related comment, also move the clearing of the QDR into its proper place. Signed-off-by: Julian Wiedmann Reviewed-by: Benjamin Block Signed-off-by: Vasily Gorbik

s390/qdio: use dev_info() in qdio_print_subchannel_info()

2021-08-18T08:01:28Z

Prefer dev_info() over a raw printk. This also adds the device and driver names into the output, so that we have: Before: qdio: 0.0.17c0 ZFCP on SC 17 using [...] After: zfcp 0.0.17c0: qdio: ZFCP on SC 17 using [...] Signed-off-by: Julian Wiedmann Reviewed-by: Benjamin Block Signed-off-by: Vasily Gorbik

s390/qdio: clean up SIGA capability tracking

2021-08-18T08:01:28Z

Don't bother with translating the SIGA-related capability bits into our own internal format, just cache the full qdioac1 field instead. Also adjust the helper macros so that they take a qdio_irq argument and can be used everywhere, instead of taking a qdio_q and then internally dereferencing the parent pointer. Signed-off-by: Julian Wiedmann Reviewed-by: Benjamin Block Signed-off-by: Vasily Gorbik

s390/qdio: remove unused sync-after-IRQ infrastructure

2021-08-18T08:01:28Z

The queue processing is fully decoupled from any preceding interrupt, so we're no longer making any use of the sync-after-IRQ HW capabilities. And as SIGA-sync is a legacy feature, there's also not much point in re-designing the driver & qdio-layer code just so that we can potentially avoid a few syncs. So just remove all the leftover code. Signed-off-by: Julian Wiedmann Reviewed-by: Benjamin Block Signed-off-by: Vasily Gorbik

s390/qdio: remove remaining tasklet & timer code

2021-07-27T07:39:18Z

Both qdio drivers have moved away from using qdio's internal tasklet and timer mechanisms for Output Queues. Rip out all the leftovers. Signed-off-by: Julian Wiedmann Reviewed-by: Benjamin Block Signed-off-by: Heiko Carstens