* [PATCH v2 1/2] nvme-pci: add module param for io queue number
@ 2018-12-24 4:12 Shan Hai
2018-12-24 4:12 ` [PATCH v2 2/2] nvme-pci: take the io_queue_number into account when setting number of io queues Shan Hai
2018-12-26 10:25 ` [PATCH v2 1/2] nvme-pci: add module param for io queue number Ming Lei
0 siblings, 2 replies; 5+ messages in thread
From: Shan Hai @ 2018-12-24 4:12 UTC (permalink / raw)
The num_possible_cpus() number of io queues by default would cause
irq vector shortage problem on a large system when hotplugging cpus,
add a module parameter to set number of io queues according to the
system configuration to fix the issue.
Below is a log of CPU hotplug failure:
[ 422.878400] WARNING: CPU: 94 PID: 0 at net/sched/sch_generic.c:334
dev_watchdog+0x228/0x22c
[ 422.878522] CPU: 94 PID: 0 Comm: swapper/94 Not tainted
[ 422.878526] task: ffff9672c0c9bd80 task.stack: ffffa7b18c964000
[ 422.878533] RIP: 0010:dev_watchdog+0x228/0x22c
[ 422.878535] RSP: 0018:ffff96b13f583e58 EFLAGS: 00010246
[ 422.878539] RAX: 0000000000000040 RBX: 0000000000000005 RCX: 000000000000083f
[ 422.878541] RDX: 0000000000000000 RSI: 00000000000000f6 RDI: 000000000000083f
[ 422.878543] RBP: ffff96b13f583e88 R08: 0000000000000000 R09: 0000000000000db6
[ 422.878546] R10: 0000000000000004 R11: 0000000000000db5 R12: ffff967131c3dbc0
[ 422.878548] R13: ffff96713a714000 R14: 000000000000005e R15: 000000000000004a
[ 422.878551] FS: 0000000000000000(0000) GS:ffff96b13f580000(0000) knlGS:0000000000000000
[ 422.878554] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 422.878556] CR2: 00007fbfcc2ca000 CR3: 0000006e7040a004 CR4: 00000000007606e0
[ 422.878559] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 422.878561] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 422.878563] PKRU: 55555554
[ 422.878564] Call Trace:
[ 422.878567] <IRQ>
[ 422.878576] ? dev_deactivate_queue.constprop.31+0x60/0x59
[ 422.878581] ? dev_deactivate_queue.constprop.31+0x60/0x59
[ 422.878590] call_timer_fn+0x3c/0x148
[ 422.878596] ? dev_deactivate_queue.constprop.31+0x60/0x59
[ 422.878601] run_timer_softirq+0x20d/0x48f
[ 422.878610] ? tick_sched_handle+0x37/0x5f
[ 422.878616] ? ktime_get+0x3e/0x95
[ 422.878627] __do_softirq+0xd9/0x28d
[ 422.878639] irq_exit+0xdf/0xe5
[ 422.878644] smp_apic_timer_interrupt+0x91/0x155
[ 422.878649] apic_timer_interrupt+0x1a2/0x1a7
[ 422.878652] </IRQ>
[ 422.878660] RIP: 0010:cpuidle_enter_state+0xdd/0x2a5
[ 422.878662] RSP: 0018:ffffa7b18c967e68 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff13
[ 422.878666] RAX: ffff96b13f5a2bc0 RBX: ffffc7b17fd85800 RCX: 000000000000001f
[ 422.878668] RDX: 0000000000000000 RSI: fffeb090fc6ad980 RDI: 0000000000000000
[ 422.878670] RBP: ffffa7b18c967ea0 R08: 0000000000008c35 R09: 000000000000c72f
[ 422.878672] R10: 000000000000c6de R11: 0000000000000018 R12: 0000000000000003
[ 422.878674] R13: 000000000000005e R14: ffffffff8756ebc0 R15: 00000062735ea43f
[ 422.878684] ? cpuidle_enter_state+0xcc/0x2a5
[ 422.878690] cpuidle_enter+0x17/0x19
[ 422.878698] call_cpuidle+0x23/0x3a
[ 422.878702] do_idle+0x172/0x1d5
[ 422.878698] call_cpuidle+0x23/0x3a
[ 422.878702] do_idle+0x172/0x1d5
[ 422.878708] cpu_startup_entry+0x73/0x75
[ 422.878718] start_secondary+0x1b9/0x208
[ 422.878736] secondary_startup_64+0xa5/0xa5
[ 422.878738] Code: 60 04 00 00 eb 8f 4c 89 ef c6 05 d8 fb eb 00 01 e8 1e f4
fc ff 89 d9 48 89 c2 4c 89 ee 48 c7 c7 b8 2f 26 87 31 c0 e8 8b c5 9b ff <0f>
0b eb bb 0f 1f 40 00 0f 1f 44 00 00 55 48 89 e5 41 57 49 89
[ 422.878758] ---[ end trace ee8cf33f8467fd66 ]---
[ 422.878761] bnxt_en 0000:46:00.0 enp70s0f0: TX timeout detected, starting
reset task!
[ 423.795776] bnxt_en 0000:46:00.0 enp70s0f0: Resp cmpl intr err msg: 0x51
[ 423.803390] bnxt_en 0000:46:00.0 enp70s0f0: hwrm_ring_free type 1 failed.
rc:ffffffff err:0
[ 424.671580] bnxt_en 0000:46:00.0 enp70s0f0: Resp cmpl intr err msg: 0x51
[ 424.679087] bnxt_en 0000:46:00.0 enp70s0f0: hwrm_ring_free type 1 failed.
rc:ffffffff err:0
[ 425.562749] bnxt_en 0000:46:00.0 enp70s0f0: Resp cmpl intr err msg: 0x51
Signed-off-by: Shan Hai <shan.hai at oracle.com>
---
v1 -> v2:
Add a cpu hotplug failure log
drivers/nvme/host/pci.c | 21 +++++++++++++++++++++
1 file changed, 21 insertions(+)
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index c33bb20..0d60451 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -64,6 +64,16 @@ MODULE_PARM_DESC(sgl_threshold,
"Use SGLs when average request segment size is larger or equal to "
"this size. Use 0 to disable SGLs.");
+static int io_queue_number_set(const char *val, const struct kernel_param *kp);
+static const struct kernel_param_ops io_queue_number_ops = {
+ .set = io_queue_number_set,
+ .get = param_get_uint,
+};
+
+static unsigned int io_queue_number = UINT_MAX;
+module_param_cb(io_queue_number, &io_queue_number_ops, &io_queue_number, 0644);
+MODULE_PARM_DESC(io_queue_number, "set io queue number, should >= 2");
+
static int io_queue_depth_set(const char *val, const struct kernel_param *kp);
static const struct kernel_param_ops io_queue_depth_ops = {
.set = io_queue_depth_set,
@@ -123,6 +133,17 @@ struct nvme_dev {
void **host_mem_desc_bufs;
};
+static int io_queue_number_set(const char *val, const struct kernel_param *kp)
+{
+ unsigned int n = 0, ret;
+
+ ret = kstrtouint(val, 10, &n);
+ if (ret != 0 || n < 2)
+ return -EINVAL;
+
+ return param_set_uint(val, kp);
+}
+
static int io_queue_depth_set(const char *val, const struct kernel_param *kp)
{
int n = 0, ret;
--
2.7.4
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH v2 2/2] nvme-pci: take the io_queue_number into account when setting number of io queues
2018-12-24 4:12 [PATCH v2 1/2] nvme-pci: add module param for io queue number Shan Hai
@ 2018-12-24 4:12 ` Shan Hai
2018-12-26 10:31 ` Ming Lei
2018-12-26 10:25 ` [PATCH v2 1/2] nvme-pci: add module param for io queue number Ming Lei
1 sibling, 1 reply; 5+ messages in thread
From: Shan Hai @ 2018-12-24 4:12 UTC (permalink / raw)
Add a wrapper around num_possible_cpus() to take the new added module
parameter account when setting number of io queues.
Signed-off-by: Shan Hai <shan.hai at oracle.com>
---
drivers/nvme/host/pci.c | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 0d60451..e359e90 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -219,6 +219,11 @@ struct nvme_iod {
struct scatterlist inline_sg[0];
};
+static inline unsigned int nvme_io_queue_number(void)
+{
+ return min_t(unsigned int, io_queue_number, num_possible_cpus());
+}
+
/*
* Check we didin't inadvertently grow the command struct
*/
@@ -241,7 +246,7 @@ static inline void _nvme_check_size(void)
static inline unsigned int nvme_dbbuf_size(u32 stride)
{
- return ((num_possible_cpus() + 1) * 8 * stride);
+ return ((nvme_io_queue_number() + 1) * 8 * stride);
}
static int nvme_dbbuf_dma_alloc(struct nvme_dev *dev)
@@ -1923,7 +1928,7 @@ static int nvme_setup_io_queues(struct nvme_dev *dev)
.pre_vectors = 1
};
- nr_io_queues = num_possible_cpus();
+ nr_io_queues = nvme_io_queue_number();
result = nvme_set_queue_count(&dev->ctrl, &nr_io_queues);
if (result < 0)
return result;
@@ -2512,7 +2517,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
if (!dev)
return -ENOMEM;
- dev->queues = kcalloc_node(num_possible_cpus() + 1,
+ dev->queues = kcalloc_node(nvme_io_queue_number() + 1,
sizeof(struct nvme_queue), GFP_KERNEL, node);
if (!dev->queues)
goto free;
--
2.7.4
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH v2 1/2] nvme-pci: add module param for io queue number
2018-12-24 4:12 [PATCH v2 1/2] nvme-pci: add module param for io queue number Shan Hai
2018-12-24 4:12 ` [PATCH v2 2/2] nvme-pci: take the io_queue_number into account when setting number of io queues Shan Hai
@ 2018-12-26 10:25 ` Ming Lei
1 sibling, 0 replies; 5+ messages in thread
From: Ming Lei @ 2018-12-26 10:25 UTC (permalink / raw)
On Mon, Dec 24, 2018@12:12:18PM +0800, Shan Hai wrote:
> The num_possible_cpus() number of io queues by default would cause
> irq vector shortage problem on a large system when hotplugging cpus,
> add a module parameter to set number of io queues according to the
> system configuration to fix the issue.
>
> Below is a log of CPU hotplug failure:
>
> [ 422.878400] WARNING: CPU: 94 PID: 0 at net/sched/sch_generic.c:334
> dev_watchdog+0x228/0x22c
> [ 422.878522] CPU: 94 PID: 0 Comm: swapper/94 Not tainted
> [ 422.878526] task: ffff9672c0c9bd80 task.stack: ffffa7b18c964000
> [ 422.878533] RIP: 0010:dev_watchdog+0x228/0x22c
> [ 422.878535] RSP: 0018:ffff96b13f583e58 EFLAGS: 00010246
> [ 422.878539] RAX: 0000000000000040 RBX: 0000000000000005 RCX: 000000000000083f
> [ 422.878541] RDX: 0000000000000000 RSI: 00000000000000f6 RDI: 000000000000083f
> [ 422.878543] RBP: ffff96b13f583e88 R08: 0000000000000000 R09: 0000000000000db6
> [ 422.878546] R10: 0000000000000004 R11: 0000000000000db5 R12: ffff967131c3dbc0
> [ 422.878548] R13: ffff96713a714000 R14: 000000000000005e R15: 000000000000004a
> [ 422.878551] FS: 0000000000000000(0000) GS:ffff96b13f580000(0000) knlGS:0000000000000000
> [ 422.878554] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 422.878556] CR2: 00007fbfcc2ca000 CR3: 0000006e7040a004 CR4: 00000000007606e0
> [ 422.878559] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> [ 422.878561] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> [ 422.878563] PKRU: 55555554
> [ 422.878564] Call Trace:
> [ 422.878567] <IRQ>
> [ 422.878576] ? dev_deactivate_queue.constprop.31+0x60/0x59
> [ 422.878581] ? dev_deactivate_queue.constprop.31+0x60/0x59
> [ 422.878590] call_timer_fn+0x3c/0x148
> [ 422.878596] ? dev_deactivate_queue.constprop.31+0x60/0x59
> [ 422.878601] run_timer_softirq+0x20d/0x48f
> [ 422.878610] ? tick_sched_handle+0x37/0x5f
> [ 422.878616] ? ktime_get+0x3e/0x95
> [ 422.878627] __do_softirq+0xd9/0x28d
> [ 422.878639] irq_exit+0xdf/0xe5
> [ 422.878644] smp_apic_timer_interrupt+0x91/0x155
> [ 422.878649] apic_timer_interrupt+0x1a2/0x1a7
> [ 422.878652] </IRQ>
> [ 422.878660] RIP: 0010:cpuidle_enter_state+0xdd/0x2a5
> [ 422.878662] RSP: 0018:ffffa7b18c967e68 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff13
> [ 422.878666] RAX: ffff96b13f5a2bc0 RBX: ffffc7b17fd85800 RCX: 000000000000001f
> [ 422.878668] RDX: 0000000000000000 RSI: fffeb090fc6ad980 RDI: 0000000000000000
> [ 422.878670] RBP: ffffa7b18c967ea0 R08: 0000000000008c35 R09: 000000000000c72f
> [ 422.878672] R10: 000000000000c6de R11: 0000000000000018 R12: 0000000000000003
> [ 422.878674] R13: 000000000000005e R14: ffffffff8756ebc0 R15: 00000062735ea43f
> [ 422.878684] ? cpuidle_enter_state+0xcc/0x2a5
> [ 422.878690] cpuidle_enter+0x17/0x19
> [ 422.878698] call_cpuidle+0x23/0x3a
> [ 422.878702] do_idle+0x172/0x1d5
> [ 422.878698] call_cpuidle+0x23/0x3a
> [ 422.878702] do_idle+0x172/0x1d5
> [ 422.878708] cpu_startup_entry+0x73/0x75
> [ 422.878718] start_secondary+0x1b9/0x208
> [ 422.878736] secondary_startup_64+0xa5/0xa5
> [ 422.878738] Code: 60 04 00 00 eb 8f 4c 89 ef c6 05 d8 fb eb 00 01 e8 1e f4
> fc ff 89 d9 48 89 c2 4c 89 ee 48 c7 c7 b8 2f 26 87 31 c0 e8 8b c5 9b ff <0f>
> 0b eb bb 0f 1f 40 00 0f 1f 44 00 00 55 48 89 e5 41 57 49 89
> [ 422.878758] ---[ end trace ee8cf33f8467fd66 ]---
> [ 422.878761] bnxt_en 0000:46:00.0 enp70s0f0: TX timeout detected, starting
> reset task!
> [ 423.795776] bnxt_en 0000:46:00.0 enp70s0f0: Resp cmpl intr err msg: 0x51
> [ 423.803390] bnxt_en 0000:46:00.0 enp70s0f0: hwrm_ring_free type 1 failed.
> rc:ffffffff err:0
> [ 424.671580] bnxt_en 0000:46:00.0 enp70s0f0: Resp cmpl intr err msg: 0x51
> [ 424.679087] bnxt_en 0000:46:00.0 enp70s0f0: hwrm_ring_free type 1 failed.
> rc:ffffffff err:0
> [ 425.562749] bnxt_en 0000:46:00.0 enp70s0f0: Resp cmpl intr err msg: 0x51
>
> Signed-off-by: Shan Hai <shan.hai at oracle.com>
> ---
>
> v1 -> v2:
> Add a cpu hotplug failure log
>
> drivers/nvme/host/pci.c | 21 +++++++++++++++++++++
> 1 file changed, 21 insertions(+)
>
> diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
> index c33bb20..0d60451 100644
> --- a/drivers/nvme/host/pci.c
> +++ b/drivers/nvme/host/pci.c
> @@ -64,6 +64,16 @@ MODULE_PARM_DESC(sgl_threshold,
> "Use SGLs when average request segment size is larger or equal to "
> "this size. Use 0 to disable SGLs.");
>
> +static int io_queue_number_set(const char *val, const struct kernel_param *kp);
> +static const struct kernel_param_ops io_queue_number_ops = {
> + .set = io_queue_number_set,
> + .get = param_get_uint,
> +};
> +
> +static unsigned int io_queue_number = UINT_MAX;
> +module_param_cb(io_queue_number, &io_queue_number_ops, &io_queue_number, 0644);
> +MODULE_PARM_DESC(io_queue_number, "set io queue number, should >= 2");
> +
I suggest to name it as 'default_queues', and 'queue_count_ops' can be
reused too, just like 'write_queues' and 'poll_queues'.
Thanks,
Ming
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v2 2/2] nvme-pci: take the io_queue_number into account when setting number of io queues
2018-12-24 4:12 ` [PATCH v2 2/2] nvme-pci: take the io_queue_number into account when setting number of io queues Shan Hai
@ 2018-12-26 10:31 ` Ming Lei
2018-12-27 1:53 ` Shan Hai
0 siblings, 1 reply; 5+ messages in thread
From: Ming Lei @ 2018-12-26 10:31 UTC (permalink / raw)
On Mon, Dec 24, 2018@12:12:19PM +0800, Shan Hai wrote:
> Add a wrapper around num_possible_cpus() to take the new added module
> parameter account when setting number of io queues.
>
> Signed-off-by: Shan Hai <shan.hai at oracle.com>
> ---
> drivers/nvme/host/pci.c | 11 ++++++++---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
> index 0d60451..e359e90 100644
> --- a/drivers/nvme/host/pci.c
> +++ b/drivers/nvme/host/pci.c
> @@ -219,6 +219,11 @@ struct nvme_iod {
> struct scatterlist inline_sg[0];
> };
>
> +static inline unsigned int nvme_io_queue_number(void)
> +{
> + return min_t(unsigned int, io_queue_number, num_possible_cpus());
> +}
> +
> /*
> * Check we didin't inadvertently grow the command struct
> */
> @@ -241,7 +246,7 @@ static inline void _nvme_check_size(void)
>
> static inline unsigned int nvme_dbbuf_size(u32 stride)
> {
> - return ((num_possible_cpus() + 1) * 8 * stride);
> + return ((nvme_io_queue_number() + 1) * 8 * stride);
> }
>
> static int nvme_dbbuf_dma_alloc(struct nvme_dev *dev)
> @@ -1923,7 +1928,7 @@ static int nvme_setup_io_queues(struct nvme_dev *dev)
> .pre_vectors = 1
> };
>
> - nr_io_queues = num_possible_cpus();
> + nr_io_queues = nvme_io_queue_number();
> result = nvme_set_queue_count(&dev->ctrl, &nr_io_queues);
> if (result < 0)
> return result;
> @@ -2512,7 +2517,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
> if (!dev)
> return -ENOMEM;
>
> - dev->queues = kcalloc_node(num_possible_cpus() + 1,
> + dev->queues = kcalloc_node(nvme_io_queue_number() + 1,
> sizeof(struct nvme_queue), GFP_KERNEL, node);
> if (!dev->queues)
> goto free;
> --
> 2.7.4
You need to make the patch against block/for-4.21, or the latest linus
tree after Linus pulls in block/for-4.21, so far not done yet.
Especially max_queue_count() and max_io_queues() are introduced, and
looks you just need to change max_io_queues().
So please fold the two into one patch, and the shape may be something
like the following:
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index d839bbe408c3..a2d1da6ca6c0 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -81,6 +81,12 @@ static const struct kernel_param_ops queue_count_ops = {
.get = param_get_int,
};
+static int default_queues;
+module_param_cb(default_queues, &queue_count_ops, &default_queues, 0644);
+MODULE_PARM_DESC(default_queues,
+ "Number of queues to use DEFAULT queue type. If not set, "
+ "num_possible_cpus() will be used.");
+
static int write_queues;
module_param_cb(write_queues, &queue_count_ops, &write_queues, 0644);
MODULE_PARM_DESC(write_queues,
@@ -254,7 +260,9 @@ static inline void _nvme_check_size(void)
static unsigned int max_io_queues(void)
{
- return num_possible_cpus() + write_queues + poll_queues;
+ int def_io_queues = default_queues ?: num_possible_cpus();
+
+ return def_io_queues + write_queues + poll_queues;
}
static unsigned int max_queue_count(void)
Thanks,
Ming
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [PATCH v2 2/2] nvme-pci: take the io_queue_number into account when setting number of io queues
2018-12-26 10:31 ` Ming Lei
@ 2018-12-27 1:53 ` Shan Hai
0 siblings, 0 replies; 5+ messages in thread
From: Shan Hai @ 2018-12-27 1:53 UTC (permalink / raw)
On 2018/12/26 ??6:31, Ming Lei wrote:
> On Mon, Dec 24, 2018@12:12:19PM +0800, Shan Hai wrote:
>> Add a wrapper around num_possible_cpus() to take the new added module
>> parameter account when setting number of io queues.
>>
>> Signed-off-by: Shan Hai <shan.hai at oracle.com>
>> ---
>> drivers/nvme/host/pci.c | 11 ++++++++---
>> 1 file changed, 8 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
>> index 0d60451..e359e90 100644
>> --- a/drivers/nvme/host/pci.c
>> +++ b/drivers/nvme/host/pci.c
>> @@ -219,6 +219,11 @@ struct nvme_iod {
>> struct scatterlist inline_sg[0];
>> };
>>
>> +static inline unsigned int nvme_io_queue_number(void)
>> +{
>> + return min_t(unsigned int, io_queue_number, num_possible_cpus());
>> +}
>> +
>> /*
>> * Check we didin't inadvertently grow the command struct
>> */
>> @@ -241,7 +246,7 @@ static inline void _nvme_check_size(void)
>>
>> static inline unsigned int nvme_dbbuf_size(u32 stride)
>> {
>> - return ((num_possible_cpus() + 1) * 8 * stride);
>> + return ((nvme_io_queue_number() + 1) * 8 * stride);
>> }
>>
>> static int nvme_dbbuf_dma_alloc(struct nvme_dev *dev)
>> @@ -1923,7 +1928,7 @@ static int nvme_setup_io_queues(struct nvme_dev *dev)
>> .pre_vectors = 1
>> };
>>
>> - nr_io_queues = num_possible_cpus();
>> + nr_io_queues = nvme_io_queue_number();
>> result = nvme_set_queue_count(&dev->ctrl, &nr_io_queues);
>> if (result < 0)
>> return result;
>> @@ -2512,7 +2517,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
>> if (!dev)
>> return -ENOMEM;
>>
>> - dev->queues = kcalloc_node(num_possible_cpus() + 1,
>> + dev->queues = kcalloc_node(nvme_io_queue_number() + 1,
>> sizeof(struct nvme_queue), GFP_KERNEL, node);
>> if (!dev->queues)
>> goto free;
>> --
>> 2.7.4
>
> You need to make the patch against block/for-4.21, or the latest linus
> tree after Linus pulls in block/for-4.21, so far not done yet.
>
> Especially max_queue_count() and max_io_queues() are introduced, and
> looks you just need to change max_io_queues().
>
> So please fold the two into one patch, and the shape may be something
> like the following:
>
Hi Ming,
Would you please go ahead and finish it? Please add me as Reported-by,
I am just fine with it, thanks for the review.
Thanks
Shan Hai
> diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
> index d839bbe408c3..a2d1da6ca6c0 100644
> --- a/drivers/nvme/host/pci.c
> +++ b/drivers/nvme/host/pci.c
> @@ -81,6 +81,12 @@ static const struct kernel_param_ops queue_count_ops = {
> .get = param_get_int,
> };
>
> +static int default_queues;
> +module_param_cb(default_queues, &queue_count_ops, &default_queues, 0644);
> +MODULE_PARM_DESC(default_queues,
> + "Number of queues to use DEFAULT queue type. If not set, "
> + "num_possible_cpus() will be used.");
> +
> static int write_queues;
> module_param_cb(write_queues, &queue_count_ops, &write_queues, 0644);
> MODULE_PARM_DESC(write_queues,
> @@ -254,7 +260,9 @@ static inline void _nvme_check_size(void)
>
> static unsigned int max_io_queues(void)
> {
> - return num_possible_cpus() + write_queues + poll_queues;
> + int def_io_queues = default_queues ?: num_possible_cpus();
> +
> + return def_io_queues + write_queues + poll_queues;
> }
>
> static unsigned int max_queue_count(void)
>
> Thanks,
> Ming
>
> _______________________________________________
> Linux-nvme mailing list
> Linux-nvme at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-nvme
>
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2018-12-27 1:53 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-12-24 4:12 [PATCH v2 1/2] nvme-pci: add module param for io queue number Shan Hai
2018-12-24 4:12 ` [PATCH v2 2/2] nvme-pci: take the io_queue_number into account when setting number of io queues Shan Hai
2018-12-26 10:31 ` Ming Lei
2018-12-27 1:53 ` Shan Hai
2018-12-26 10:25 ` [PATCH v2 1/2] nvme-pci: add module param for io queue number Ming Lei
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox