linux-dev/drivers/s390/net/qeth_core.h, branch linus/master

qeth: remove a copy of the NAPI_POLL_WEIGHT define

2022-04-29T10:56:42Z

Defining local versions of NAPI_POLL_WEIGHT with the same values in the drivers just makes refactoring harder. Signed-off-by: Jakub Kicinski Acked-by: Alexandra Winter Signed-off-by: David S. Miller

s390/qeth: fine-tune .ndo_select_queue()

2021-12-08T06:01:02Z

Avoid a conditional branch for L2 devices when selecting the TX queue, and have shared logic for OSA devices. Signed-off-by: Julian Wiedmann Signed-off-by: Alexandra Winter Signed-off-by: Jakub Kicinski

s390/qeth: add __printf format attribute to qeth_dbf_longtext

2021-10-25T12:58:19Z

Allow the compiler to recognize and check format strings and parameters. As reported with allmodconfig and W=1: drivers/s390/net/qeth_core_main.c: In function ‘qeth_dbf_longtext’: drivers/s390/net/qeth_core_main.c:6190:9: error: function ‘qeth_dbf_longtext’ might be a candidate for ‘gnu_printf’ format attribute [-Werror=suggest-attribute=format] 6190 | vsnprintf(dbf_txt_buf, sizeof(dbf_txt_buf), fmt, args); | ^~~~~~~~~ Acked-by: Julian Wiedmann Signed-off-by: Heiko Carstens Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller

s390/qeth: don't keep track of Input Queue count

2021-10-25T12:58:19Z

The only actual user of qdio.no_input_queues is qeth_qdio_establish(), and there we already have full awareness of the current Input Queue configuration (1 RX queue, plus potentially 1 TX Completion queue). So avoid this state tracking, and the ambiguity it brings with it. Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller

s390/qeth: remove .do_ioctl() callback from driver discipline

2021-10-25T12:58:19Z

With commit 18787eeebd71 ("qeth: use ndo_siocdevprivate") this callback is now actually used to handle transport mode-specific _private_ ioctls. We only have such ioctls for L3 devices. So wire up a L3-specific .ndo_siocdevprivate() callback that handles those ioctls, and defers to the core qeth_siocdevprivate() for all other private ioctls. This takes the discipline one step closer to its original purpose of providing an internal extension for the qeth_core_ccwgroup_driver. Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller

s390/qeth: Fix deadlock in remove_discipline

2021-09-22T03:02:24Z

Problem: qeth_close_dev_handler is a worker that tries to acquire card->discipline_mutex via drv->set_offline() in ccwgroup_set_offline(). Since commit b41b554c1ee7 ("s390/qeth: fix locking for discipline setup / removal") qeth_remove_discipline() is called under card->discipline_mutex and cancels the work and waits for it to finish. STOPLAN reception with reason code IPA_RC_VEPA_TO_VEB_TRANSITION is the only situation that schedules close_dev_work. In that situation scheduling qeth recovery will also result in an offline interface, when resetting the isolation mode fails, if the external switch is still set to VEB. And since commit 0b9902c1fcc5 ("s390/qeth: fix deadlock during recovery") qeth recovery does not aquire card->discipline_mutex anymore. So we accept the longer pathlength of qeth_schedule_recovery in this error situation and re-use the existing function. As a side-benefit this changes the hwtrap to behave like during recovery instead of like during a user-triggered set_offline. Fixes: b41b554c1ee7 ("s390/qeth: fix locking for discipline setup / removal") Signed-off-by: Alexandra Winter Acked-by: Julian Wiedmann Signed-off-by: Julian Wiedmann Signed-off-by: Jakub Kicinski

qeth: use ndo_siocdevprivate

2021-07-27T19:11:44Z

qeth has both standard MII ioctls and custom SIOCDEVPRIVATE ones, all of which work correctly with compat user space. Move the private ones over to the new ndo_siocdevprivate callback. Cc: Julian Wiedmann Cc: Karsten Graul Cc: Heiko Carstens Cc: Vasily Gorbik Cc: Christian Borntraeger Cc: linux-s390@vger.kernel.org Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller

s390/qeth: clean up device_type management

2021-07-20T13:17:06Z

qeth uses three device_type structs - a generic one, and one for each sub-driver (which is used for fixed-layer devices only). Instead of exporting these device_types back&forth between the driver's modules, make all the logic self-contained within the sub-drivers. On disc->setup() they either install their own device_type, or add the sysfs attributes that are missing in the generic device_type. Later on disc->remove() these attributes are removed again from any device that has the generic device_type. Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller

s390/qeth: remove OSN support

2021-07-20T13:17:06Z

Commit fb64de1bc36c ("s390/qeth: phase out OSN support") spelled out why the OSN support in qeth is in a bad shape, and put any remaining interested parties on notice to speak up before it gets ripped out. It's 2021 now, so make true on that promise and remove all the OSN-specific parts from qeth. This also means that we no longer need to export various parts of the cmd & data path internals to the L2 driver. Signed-off-by: Julian Wiedmann Reviewed-by: Alexandra Winter Signed-off-by: David S. Miller

s390/qeth: shrink TX buffer struct

2021-06-11T19:49:15Z

Convert the large boolean array into a bitmap, this substantially reduces the struct's size. While at it also clarify the naming. Signed-off-by: Julian Wiedmann Signed-off-by: David S. Miller