[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <1562606652-7618-1-git-send-email-alan.mikhak@sifive.com>
Date: Mon, 8 Jul 2019 10:24:12 -0700
From: Alan Mikhak <alan.mikhak@...ive.com>
To: linux-nvme@...ts.infradead.org, linux-kernel@...r.kernel.org,
kbusch@...nel.org, axboe@...com, hch@....de, sagi@...mberg.me,
palmer@...ive.com, paul.walmsley@...ive.com
Cc: Alan Mikhak <alan.mikhak@...ive.com>
Subject: [PATCH] nvme-pci: No IRQ map for read if no read queue allocated
From: Alan Mikhak <alan.mikhak@...ive.com>
nvme_dev_add() assumes a read queue is always allocated.
That may not be the case on single-interrupt systems for
which pci_assign_irq() would report runtime IRQ mapping
not provided by arch.
This patch makes sure nvme_dev_add() only requests an
IRQ mapping for read queues if at least one read queue
is being allocted.
nvme_pci_map_queues() will later on ignore the unnecessary
mapping request should nvme_dev_add() request such an IRQ
mapping even though no read queues are being allocated.
However, nvme_dev_add() can avoid making the request by
checking the number of read queues without assuming. This
would bring it more in line with nvme_setup_irqs() and
nvme_calc_irq_sets().
Signed-off-by: Alan Mikhak <alan.mikhak@...ive.com>
---
drivers/nvme/host/pci.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 524d6bd6d095..86262ebe6fff 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -2263,7 +2263,9 @@ static int nvme_dev_add(struct nvme_dev *dev)
if (!dev->ctrl.tagset) {
dev->tagset.ops = &nvme_mq_ops;
dev->tagset.nr_hw_queues = dev->online_queues - 1;
- dev->tagset.nr_maps = 2; /* default + read */
+ dev->tagset.nr_maps = 1; /* default */
+ if (dev->io_queues[HCTX_TYPE_READ])
+ dev->tagset.nr_maps++;
if (dev->io_queues[HCTX_TYPE_POLL])
dev->tagset.nr_maps++;
dev->tagset.timeout = NVME_IO_TIMEOUT;
--
2.7.4
Powered by blists - more mailing lists