- 07 Nov, 2018 40 commits
-
-
Jens Axboe authored
Doesn't do anything right now, but it's needed as a prep patch to get the interfaces right. While in there, correct the blk_mq_map_queue() CPU type to an unsigned int. Reviewed-by:
Hannes Reinecke <hare@suse.com> Reviewed-by:
Keith Busch <keith.busch@intel.com> Signed-off-by:
Jens Axboe <axboe@kernel.dk>
-
Jens Axboe authored
This is in preparation for allowing multiple sets of maps per queue, if so desired. Reviewed-by:
Bart Van Assche <bvanassche@acm.org> Reviewed-by:
-
Jens Axboe authored
It's just a pointer to set->mq_map, use that instead. Move the assignment a bit earlier, so we always know it's valid. Reviewed-by:
Christoph Hellwig <hch@lst.de> Reviewed-by:
-
Jens Axboe authored
Merge branch 'irq/for-block' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip into for-4.21/block Pull in the irq affinity commits, that are staged through Thomas's tree. * 'irq/for-block' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: genirq/affinity: Add support for allocating interrupt sets genirq/affinity: Pass first vector to __irq_build_affinity_masks() genirq/affinity: Move two stage affinity spreading into a helper function genirq/affinity: Spread IRQs to all available NUMA nodes
-
Jens Axboe authored
This was used for completion placement for the legacy path, but for mq we have rq->mq_ctx->cpu for that. Add a helper to get the request CPU assignment, as the mq_ctx type is private to blk-mq. Reviewed-by:
Omar Sandoval <osandov@fb.com> Signed-off-by:
-
Jens Axboe authored
With the legacy path gone, all we do is funnel it through the mq_ops->complete() operation. Tested-by:
Ming Lei <ming.lei@redhat.com> Reviewed-by:
-
Jens Axboe authored
No point in hiding what this does, just open code it in the one spot where we are still using it. Reviewed-by:
-
Jens Axboe authored
This is dead code, any queue reaching this part has mq_ops attached. Reviewed-by:
-
Jens Axboe authored
It'll always be false at this point, just remove it. Reviewed-by:
-
Jens Axboe authored
Reviewed-by:
-
Jens Axboe authored
It's now dead code, nobody uses it. Reviewed-by:
-
Jens Axboe authored
Get rid of the special bsg job fn and timeout handler, move them into a private bsg_set instead. Mostly from Christoph, with fixes for error handling and cleanups. Reviewed-by:
-
Jens Axboe authored
The only user of legacy timing now is BSG, which is invoked from the mq timeout handler. Kill the legacy code, and rename the q->rq_timed_out_fn to q->bsg_job_timeout_fn. Reviewed-by:
-
Jens Axboe authored
Now there's no difference between blk_put_request() and __blk_put_request() anymore, get rid of the underscore version and convert the few callers. Reviewed-by:
-
Jens Axboe authored
This is a remnant of when we had ops for both SQ and MQ schedulers. Now it's just MQ, so get rid of the union. Reviewed-by:
-
Jens Axboe authored
This removes a bunch of core and elevator related code. On the core front, we remove anything related to queue running, draining, initialization, plugging, and congestions. We also kill anything related to request allocation, merging, retrieval, and completion. Remove any checking for single queue IO schedulers, as they no longer exist. This means we can also delete a bunch of code related to request issue, adding, completion, etc - and all the SQ related ops and helpers. Also kill the load_default_modules(), as all that did was provide for a way to load the default single queue elevator. Tested-by:
-
Jens Axboe authored
Retain the deadline documentation, as that carries over to mq-deadline as well. Tested-by:
-
Jens Axboe authored
Now that blk_flush_queue_rq() always returns false, we can remove that return value. That bubbles through the stack, allowing us to remove a bunch of state tracking around it. Tested-by:
-
Jens Axboe authored
Reviewed-by:
-
Jens Axboe authored
It's now unused, kill it. Reviewed-by:
-
Jens Axboe authored
We only support mq devices now. Reviewed-by:
-
Jens Axboe authored
Everything is blk-mq at this point, so it doesn't make any sense to have this option available as it does nothing. Reviewed-by:
-
Jens Axboe authored
It's now unused. Reviewed-by:
-
Jens Axboe authored
Requires a few changes to the FC transport class as well. Cc: linux-scsi@vger.kernel.org Reviewed-by:
Johannes Thumshirn <jthumshirn@suse.de> Tested-by:
Benjamin Block <bblock@linux.vnet.ibm.com> Tested-by:
-
Jens Axboe authored
All drivers do unregister + cleanup, provide a helper for that. Cc: linux-scsi@vger.kernel.org Reviewed-by:
-
Jens Axboe authored
This will ease in the conversion to blk-mq, where we can't set a timeout handler after queue init. Cc: Johannes Thumshirn <jthumshirn@suse.de> Cc: linux-scsi@vger.kernel.org Reviewed-by:
-
Jens Axboe authored
Since e443343e we haven't had a request_fn attached to this driver, hence any code inside an if (q->request_fn) is unreachable. Fixes: e443343e ("s390/dasd: blk-mq conversion") [sth: Keep and fix the dasd_info->chanq_len counter.] Reviewed-by:
Jan Hoeppner <hoeppner@linux.ibm.com> Signed-off-by:
Stefan Haberland <sth@linux.ibm.com> Tested-by:
-
Jens Axboe authored
Nobody is using the legacy path for blk_lld_busy() anymore, remove it. Reviewed-by:
-
Jens Axboe authored
This removes the legacy (non-mq) IO path for SCSI. Cc: linux-scsi@vger.kernel.org Acked-by:
Himanshu Madhani <himanshu.madhani@cavium.com> Reviewed-by:
Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by:
-
Jens Axboe authored
Only the SCSI legacy path provides a way to check if target is currently busy, provide the same for the MQ path. Cc: linux-scsi@vger.kernel.org Reviewed-by:
-
Jens Axboe authored
We'll hook into this from blk_lld_busy(), allowing blk-mq to also return whether or not a given queue currently has requests in progress. Reviewed-by:
-
Jens Axboe authored
Reviewed-by:
-
Jens Axboe authored
We don't do anything with it, that's just the legacy path. Reviewed-by:
-
Jens Axboe authored
ide-disk and ide-cd tested as working just fine, ide-tape and ide-floppy haven't. But the latter don't require changes, so they should work without issue. Add helper function to insert a request from a work queue, since we cannot invoke the blk-mq request insertion from IRQ context. Cc: David Miller <davem@davemloft.net> Reviewed-by:
-
Jens Axboe authored
Straight forward conversion, there's room for improvement. Reviewed-by:
-
Jens Axboe authored
Straight forward conversion, room for optimization in how everything is punted to a work queue. Also looks plenty racy all over the map, with the state changes. I fixed a bunch of them up while doing the conversion, but there are surely more. Cc: Maxim Levitsky <maximlevitsky@gmail.com> Reviewed-by:
-
Jens Axboe authored
Convert from the old request_fn style driver to blk-mq. Cc: David Miller <davem@davemloft.net> Reviewed-by:
-
Masato Suzuki authored
Allow the creation of conventional zones by adding the zone_nr_conv configuration attribute. This new attribute is used only for zoned devices and indicates the number of conventional zones to create. The default value is 0. Since host-managed zoned block devices must always have at least one sequential zone, if the value of zone_nr_conv is larger than or equal to the total number of zones of the device nr_zones, zone_nr_conv is automatically changed to nr_zones - 1. Reviewed-by:
Matias Bjorling <matias.bjorling@wdc.com> Reviewed-by:
Damien Le Moal <damien.lemoal@wdc.com> Signed-off-by:
Masato Suzuki <masato.suzuki@wdc.com> Signed-off-by:
-
Linus Torvalds authored
Merge tag 'hwmon-for-v4.20-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging Pull hwmon fixes from Guenter Roeck: - Remove bogus __init annotations in ibmpowernv driver - Fix double-free in error handling of __hwmon_device_register() * tag 'hwmon-for-v4.20-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging: hwmon: (ibmpowernv) Remove bogus __init annotations hwmon: (core) Fix double-free in __hwmon_device_register()
-
git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-socLinus Torvalds authored
Pull ARM SoC fixes from Olof Johansson: "A few more fixes that have come in, and one revert of a previous fix. I was a bit too trigger happy to enable PREEMPT on multi_v7_defconfig, and it ended up regressing at least BeagleBone XM boards. While we get that debugged for next merge window, let's disable it again. Beyond that: - Stratix change to fix multicast filtering - Minor DT fixes for Renesas and i.MX - Ethernet fix for a Renesas board (switching main interfaces) - Ethernet phy regulator fix for i.MX6SX" * tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: arm64: dts: stratix10: fix multicast filtering ARM: defconfig: Disable PREEMPT again on multi_v7 arm64: dts: renesas: condor: switch from EtherAVB to GEther dt-bindings: arm: Fix RZ/G2E part number arm64: dts: renesas: r8a7795: add missing dma-names on hscif2 ARM: dts: imx6sx-sdb: Fix enet phy regulator ARM: dts: fsl: Fix improperly quoted stdout-path values ARM: dts: imx6sll: fix typo for fsl,imx6sll-i2c node
-
