Re: [RFC/RFT PATCH] scsi: pm8001: Expose HW queues for pm80xx hw

From: John Garry
Date: Mon Feb 01 2021 - 12:19:24 EST


On 01/02/2021 17:08, Viswas.G@xxxxxxxxxxxxx wrote:
Hi John,

AFAIK, there is no such restrictions. Any queue can be used for internal/external commands.


ok, understood.

BTW, to see even more performance improvement, it would be good to use request->tag for ccb tag, rather that the LLDD manage this itself. I mentioned this perviously elsewhere. Do you plan to make that change? hisi_sas and megaraid sas are examples of drivers who do this.

Thanks,
John

Regards,
Viswas G

-----Original Message-----
From: John Garry <john.garry@xxxxxxxxxx>
Sent: Monday, February 1, 2021 4:53 PM
To: Viswas G - I30667 <Viswas.G@xxxxxxxxxxxxx>; jejb@xxxxxxxxxxxxx;
martin.petersen@xxxxxxxxxx; akshatzen@xxxxxxxxxx; Ruksar Devadi -
I52327 <Ruksar.devadi@xxxxxxxxxxxxx>; radha@xxxxxxxxxx;
bjashnani@xxxxxxxxxx; vishakhavc@xxxxxxxxxx;
jinpu.wang@xxxxxxxxxxxxxxx; Ashokkumar N - X53535
<Ashokkumar.N@xxxxxxxxxxxxx>
Cc: linux-scsi@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; hare@xxxxxxx;
kashyap.desai@xxxxxxxxxxxx; ming.lei@xxxxxxxxxx
Subject: Re: [RFC/RFT PATCH] scsi: pm8001: Expose HW queues for pm80xx
hw

EXTERNAL EMAIL: Do not click links or open attachments unless you know the
content is safe

On 31/01/2021 07:19, Viswas.G@xxxxxxxxxxxxx wrote:
Thanks Johns.

We could see a kernel crash while testing this patch.

Thanks for testing.


[ 246.724632] scsi host10: pm80xx
[ 248.005258] sas: Enter sas_scsi_recover_host busy: 0 failed: 0 [
248.168973] BUG: kernel NULL pointer dereference, address:
0000000000000110 [ 248.175926] #PF: supervisor read access in kernel
mode [ 248.181065] #PF: error_code(0x0000) - not-present page [
248.186196] PGD 0 P4D 0 [ 248.188736] Oops: 0000 [#1] SMP PTI
[ 248.192230] CPU: 10 PID: 77 Comm: kworker/u26:2 Kdump: loaded
Tainted: G S OE 5.11.0-rc3 #2
[ 248.201614] Hardware name: Supermicro Super Server/X10DRi-LN4+,
BIOS 3.1 06/08/2018 [ 248.209258] Workqueue: events_unbound
async_run_entry_fn [ 248.214571] RIP:
0010:pm80xx_chip_sata_req+0x7f/0x5e0 [pm80xx] [ 248.220413] Code: c1
7c c1 e9 03 4d 8b ac 24 80 01 00 00 48 c7 44 24 14 00 00 00 00 89 04
24 31 c0 48 c7 84 24 88 00 00 00 00 00 00 00 f3 48 ab <48> 8b ba 10 01
00 00 e8 35 35 c6 ef c1 e8 10 89 44 24 04 0f b6 43 [ 248.239157] RSP:
0018:ffffb98d834979f0 EFLAGS: 00010046 [ 248.244384] RAX:
0000000000000000 RBX: ffff9523c321c000 RCX: 0000000000000000 [
248.251516] RDX: 0000000000000000 RSI: ffff952450720048 RDI:
ffffb98d83497a80 [ 248.258641] RBP: ffff9523c7420000 R08:
0000000000000100 R09: 0000000000000001 [ 248.265764] R10:
0000000000000001 R11: ffff9523c9e40000 R12: ffff9523ca1c3600 [
248.272887] R13: ffff9523c9e40000 R14: ffffb98d83497a04 R15:
ffff952450720048 [ 248.280013] FS: 0000000000000000(0000)
GS:ffff9527afd00000(0000) knlGS:0000000000000000 [ 248.288090] CS: 0010
DS: 0000 ES: 0000 CR0: 0000000080050033 [ 248.293826] CR2:
0000000000000110 CR3: 000000029ac10001 CR4: 00000000001706e0 [
248.300952] Call Trace:

I think that the problem here is that ata_queued_cmd->scmd is NULL for the
ata internal command.

Please try this fix:

---->8-----

@@ -4451,7 +4451,7 @@ static int pm80xx_chip_sata_req(struct
pm8001_hba_info *pm8001_ha,
struct scsi_cmnd *scmd = qc->scsicmd;
u32 tag = ccb->ccb_tag;
int ret;
- u32 q_index, blk_tag;
+ u32 q_index = 0, blk_tag;
struct sata_start_req sata_cmd;
u32 hdr_tag, ncg_tag = 0;
u64 phys_addr, start_addr, end_addr; @@ -4463,8 +4463,10 @@ static
int pm80xx_chip_sata_req(struct pm8001_hba_info *pm8001_ha,
u32 opc = OPC_INB_SATA_HOST_OPSTART;
memset(&sata_cmd, 0, sizeof(sata_cmd));

- blk_tag = blk_mq_unique_tag(scmd->request);
- q_index = blk_mq_unique_tag_to_hwq(blk_tag);
+ if (scmd) {
+ blk_tag = blk_mq_unique_tag(scmd->request);
+ q_index = blk_mq_unique_tag_to_hwq(blk_tag);
+ }
circularQ = &pm8001_ha->inbnd_q_tbl[q_index];

----8<-----

You may need similar for other dispatch paths also - like smp - but I don't
think that you will.

BTW, do you know if there is actually a limitation on the HW that queue
#0 is used for all "internal" IO (mpi), like phy control operation? Jack?

Thanks,
John


[ 248.303402] pm8001_task_exec.isra.9+0x2a4/0x460 [pm80xx] [
248.308805] sas_ata_qc_issue+0x187/0x220 [libsas] [ 248.313607]
ata_qc_issue+0x107/0x1e0 [libata] [ 248.318069]
ata_exec_internal_sg+0x2c8/0x580 [libata] [ 248.323217]
ata_exec_internal+0x5f/0x90 [libata] [ 248.327931]
ata_dev_read_id+0x306/0x480 [libata] [ 248.332647]
ata_eh_recover+0x7ea/0x12a0 [libata] [ 248.337369] ?
vprintk_emit+0x114/0x220 [ 248.341208] ? sas_ata_sched_eh+0x60/0x60
[libsas] [ 248.346002] ? sas_ata_prereset+0x50/0x50 [libsas] [
248.350795] ? printk+0x58/0x6f [ 248.353941] ?
sas_ata_sched_eh+0x60/0x60 [libsas] [ 248.358733] ?
sas_ata_prereset+0x50/0x50 [libsas] [ 248.363525]
ata_do_eh+0x40/0xb0 [libata] [ 248.367556]
ata_scsi_port_error_handler+0x354/0x770 [libata] [ 248.373318]
async_sas_ata_eh+0x44/0x7b [libsas] [ 248.377938]
async_run_entry_fn+0x39/0x160 [ 248.382040]
process_one_work+0x1cb/0x360 [ 248.386050]
worker_thread+0x30/0x370
[ 248.389706] ? processe_work+0x360/0x360 [ 248.393884]
kthread+0x116/0x130 [ 248.397116] ? kthread_park+0x80/0x80 [
248.400773] ret_from_fork+0x22/0x30 [ 248.404355] Modules linked in:
pm80xx(OE) libsas scsi_transport_sas xt_CHECKSUM xt_MASQUERADE
xt_conntrack ipt_REJECT nf_reject_ipv4 nft_compat nft_counter
nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 nf_tables
nfnetlink tun bridge stp llc rfkill sunrpc vfat fat intel_rapl_msr
intel_rapl_common sb_edac x86_pkg_temp_thermal intel_powerclamp
coretemp kvm_intel kvm irqbypass joydev crct10dif_pclmul crc32_pclmul
ghash_clmulni_intel ipmi_ssif iTCO_wdt iTCO_vendor_support mei_me rapl
i2c_i801 intel_cstate mei acpi_ipmi lpc_ich intel_uncore pcspkr ipmi_si
i2c_smbus ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad
ioatdma ip_tables xfs libcrc32c sr_mod sd_mod cdrom t10_pi sg ast
drm_vram_helper drm_kms_helper syscopyarea igb sysfillrect sysimgblt
fb_sys_fops drm_ttm_helper ttm ahci libahci dca drm crc32c_intel libata
i2c_algo_bit wmi dm_mirror dm_region_hash dm_log dm_mod fuse [
248.483431] CR2: 0000000000000110
[ 0.000000] Linux version 5.11.0-rc3 (root@localhost.localdomain) (gcc
(GCC) 8.3.1 20191121 (Red Hat 8.3.1-5), GNU ld version 2.30-79.el8) #2 SMP
Mon Jan 25 23:56:12 IST 2021
[ 0.000000] Command line: elfcorehr=0x45000000
BOOT_IMAGE=(hd14,gpt2)/vmlinuz-5.11.0-rc3 ro
resume=/dev/mapper/rhel-swap rhgb console=ttyS1,115200 loglevel=7
irqpoll nr_cpus=1 reset_devices cgroup_disable=memory mce=off numa=off
udev.children-max=2 panic=10 rootflags=nofail acpi_no_memhotplug
transparent_hugepage=never nokaslr novmcoredd hest_disable
disable_cpu_apicid=0


-----Original Message-----
From: John Garry <john.garry@xxxxxxxxxx>
Sent: Tuesday, January 5, 2021 4:47 PM
To: jejb@xxxxxxxxxxxxx; martin.petersen@xxxxxxxxxx;
akshatzen@xxxxxxxxxx; Viswas G - I30667 <Viswas.G@xxxxxxxxxxxxx>;
Ruksar Devadi - I52327 <Ruksar.devadi@xxxxxxxxxxxxx>;
radha@xxxxxxxxxx; bjashnani@xxxxxxxxxx; vishakhavc@xxxxxxxxxx;
jinpu.wang@xxxxxxxxxxxxxxx; Ashokkumar N - X53535
<Ashokkumar.N@xxxxxxxxxxxxx>
Cc: linux-scsi@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx;
hare@xxxxxxx; kashyap.desai@xxxxxxxxxxxx; ming.lei@xxxxxxxxxx;
John
Garry <john.garry@xxxxxxxxxx>
Subject: [RFC/RFT PATCH] scsi: pm8001: Expose HW queues for pm80xx
hw

EXTERNAL EMAIL: Do not click links or open attachments unless you
know the content is safe

In commit 05c6c029a44d ("scsi: pm80xx: Increase number of supported
queues"), support for 80xx chip was improved by enabling multiple HW
queues.

In this, like other SCSI MQ HBA drivers, the HW queues were not
exposed to upper layer, and instead the driver managed the queues
internally.

However, this management duplicates blk-mq code. In addition, the HW
queue management is sub-optimal for a system where the number of
CPUs
exceeds the HW queues - this is because queues are selected in a
round- robin fashion, when it would be better to make adjacent CPUs
submit on the same queue. And finally, the affinity of the completion
queue interrupts is not set to mirror the cpu<->HQ queue mapping, which
is suboptimal.

As such, for when MSIX is supported, expose HW queues to upper layer.
Flag PCI_IRQ_AFFINITY is set for allocating the MSIX vectors to
automatically assign affinity for the completion queue interrupts.

Signed-off-by: John Garry <john.garry@xxxxxxxxxx>

---
I sent as an RFC/RFT as I have no HW to test. In addition, since HW
queue
#0 is used always for internal commands (like in send_task_abort()),
if all CPUs associated with HW queue #0 are offlined, the interrupt
for that queue will be shutdown, and no CPUs would be available to
service any internal commands completion. To solve that, we need [0]
merged first and switch over to use the new API. But we can still test
performance in the meantime.

I assume someone else is making the change to use the request tag for
IO tag management.

[0] https://lore.kernel.org/linux-scsi/47ba045e-a490-198b-1744-
529f97192d3b@xxxxxxx/

diff --git a/drivers/scsi/pm8001/pm8001_init.c
b/drivers/scsi/pm8001/pm8001_init.c
index ee2de177d0d0..73479803a23e 100644
--- a/drivers/scsi/pm8001/pm8001_init.c
+++ b/drivers/scsi/pm8001/pm8001_init.c
@@ -81,6 +81,15 @@ LIST_HEAD(hba_list);

struct workqueue_struct *pm8001_wq;

+static int pm8001_map_queues(struct Scsi_Host *shost) {
+ struct sas_ha_struct *sha = SHOST_TO_SAS_HA(shost);
+ struct pm8001_hba_info *pm8001_ha = sha->lldd_ha;
+ struct blk_mq_queue_map *qmap =
+&shost->tag_set.map[HCTX_TYPE_DEFAULT];
+
+ return blk_mq_pci_map_queues(qmap, pm8001_ha->pdev, 0); }
+
/*
* The main structure which LLDD must register for scsi core.
*/
@@ -106,6 +115,7 @@ static struct scsi_host_template pm8001_sht = {
#ifdef CONFIG_COMPAT
.compat_ioctl = sas_ioctl,
#endif
+ .map_queues = pm8001_map_queues,
.shost_attrs = pm8001_host_attrs,
.track_queue_depth = 1,
};
@@ -923,9 +933,8 @@ static int pm8001_configure_phy_settings(struct
pm8001_hba_info *pm8001_ha) static u32 pm8001_setup_msix(struct
pm8001_hba_info *pm8001_ha) {
u32 number_of_intr;
- int rc, cpu_online_count;
+ int rc;
unsigned int allocated_irq_vectors;
-
/* SPCv controllers supports 64 msi-x */
if (pm8001_ha->chip_id == chip_8001) {
number_of_intr = 1;
@@ -933,16 +942,15 @@ static u32 pm8001_setup_msix(struct
pm8001_hba_info *pm8001_ha)
number_of_intr = PM8001_MAX_MSIX_VEC;
}

- cpu_online_count = num_online_cpus();
- number_of_intr = min_t(int, cpu_online_count, number_of_intr);
- rc = pci_alloc_irq_vectors(pm8001_ha->pdev, number_of_intr,
- number_of_intr, PCI_IRQ_MSIX);
+ /* Use default affinity descriptor, which spreads *all* vectors */
+ rc = pci_alloc_irq_vectors(pm8001_ha->pdev, 1,
+ number_of_intr, PCI_IRQ_MSIX |
+ PCI_IRQ_AFFINITY);
allocated_irq_vectors = rc;
if (rc < 0)
return rc;

/* Assigns the number of interrupts */
- number_of_intr = min_t(int, allocated_irq_vectors, number_of_intr);
+ number_of_intr = allocated_irq_vectors;
pm8001_ha->number_of_intr = number_of_intr;

/* Maximum queue number updating in HBA structure */ @@
-1113,6
+1121,16 @@ static int pm8001_pci_probe(struct pci_dev *pdev,
if (rc)
goto err_out_enable;

+ if (pm8001_ha->number_of_intr > 1) {
+ shost->nr_hw_queues = pm8001_ha->number_of_intr;
+ /*
+ * For now, ensure we're not sent too many commands by
setting
+ * host_tagset. This is also required if we start using request
+ * tag.
+ */
+ shost->host_tagset = 1;
+ }
+
rc = scsi_add_host(shost, &pdev->dev);
if (rc)
goto err_out_ha_free; diff --git
a/drivers/scsi/pm8001/pm8001_sas.h
b/drivers/scsi/pm8001/pm8001_sas.h
index f2c8cbad3853..74bc6fed693e 100644
--- a/drivers/scsi/pm8001/pm8001_sas.h
+++ b/drivers/scsi/pm8001/pm8001_sas.h
@@ -55,6 +55,8 @@
#include <scsi/scsi_tcq.h>
#include <scsi/sas_ata.h>
#include <linux/atomic.h>
+#include <linux/blk-mq.h>
+#include <linux/blk-mq-pci.h>
#include "pm8001_defs.h"

#define DRV_NAME "pm80xx"
diff --git a/drivers/scsi/pm8001/pm80xx_hwi.c
b/drivers/scsi/pm8001/pm80xx_hwi.c
index 6772b0924dac..31d65ce91e7d 100644
--- a/drivers/scsi/pm8001/pm80xx_hwi.c
+++ b/drivers/scsi/pm8001/pm80xx_hwi.c
@@ -4299,12 +4299,13 @@ static int pm80xx_chip_ssp_io_req(struct
pm8001_hba_info *pm8001_ha,
struct domain_device *dev = task->dev;
struct pm8001_device *pm8001_dev = dev->lldd_dev;
struct ssp_ini_io_start_req ssp_cmd;
+ struct scsi_cmnd *scmd = task->uldd_task;
u32 tag = ccb->ccb_tag;
int ret;
u64 phys_addr, start_addr, end_addr;
u32 end_addr_high, end_addr_low;
struct inbound_queue_table *circularQ;
- u32 q_index, cpu_id;
+ u32 blk_tag, q_index;
u32 opc = OPC_INB_SSPINIIOSTART;
memset(&ssp_cmd, 0, sizeof(ssp_cmd));
memcpy(ssp_cmd.ssp_iu.lun, task->ssp_task.LUN, 8); @@
-4323,8
+4324,8 @@ static int pm80xx_chip_ssp_io_req(struct pm8001_hba_info
*pm8001_ha,
ssp_cmd.ssp_iu.efb_prio_attr |= (task->ssp_task.task_attr & 7);
memcpy(ssp_cmd.ssp_iu.cdb, task->ssp_task.cmd->cmnd,
task->ssp_task.cmd->cmd_len);
- cpu_id = smp_processor_id();
- q_index = (u32) (cpu_id) % (pm8001_ha->max_q_num);
+ blk_tag = blk_mq_unique_tag(scmd->request);
+ q_index = blk_mq_unique_tag_to_hwq(blk_tag);
circularQ = &pm8001_ha->inbnd_q_tbl[q_index];

/* Check if encryption is set */ @@ -4446,9 +4447,11 @@
static int pm80xx_chip_sata_req(struct pm8001_hba_info *pm8001_ha,
struct sas_task *task = ccb->task;
struct domain_device *dev = task->dev;
struct pm8001_device *pm8001_ha_dev = dev->lldd_dev;
+ struct ata_queued_cmd *qc = task->uldd_task;
+ struct scsi_cmnd *scmd = qc->scsicmd;
u32 tag = ccb->ccb_tag;
int ret;
- u32 q_index, cpu_id;
+ u32 q_index, blk_tag;
struct sata_start_req sata_cmd;
u32 hdr_tag, ncg_tag = 0;
u64 phys_addr, start_addr, end_addr; @@ -4459,8 +4462,9 @@
static int pm80xx_chip_sata_req(struct pm8001_hba_info *pm8001_ha,
unsigned long flags;
u32 opc = OPC_INB_SATA_HOST_OPSTART;
memset(&sata_cmd, 0, sizeof(sata_cmd));
- cpu_id = smp_processor_id();
- q_index = (u32) (cpu_id) % (pm8001_ha->max_q_num);
+
+ blk_tag = blk_mq_unique_tag(scmd->request);

Here the scsi command is NULL.

if(scmd) {
blk_tag = blk_mq_unique_tag(scmd->request);
q_index = blk_mq_unique_tag_to_hwq(blk_tag);
} else {
q_index = 0;
}

With this change, we have started our testing. We will update you the
result soon.

+ q_index = blk_mq_unique_tag_to_hwq(blk_tag);
circularQ = &pm8001_ha->inbnd_q_tbl[q_index];

if (task->data_dir == DMA_NONE) {
--
2.26.2

.