All of lore.kernel.org
 help / color / mirror / Atom feed
From: <gregkh@linuxfoundation.org>
To: alexander.levin@verizon.com, gregkh@linuxfoundation.org,
	jthumshirn@suse.de, manoj@linux.vnet.ibm.com,
	martin.petersen@oracle.com, mrochs@linux.vnet.ibm.com,
	stable@vger.kernel.org, ukrishn@linux.vnet.ibm.com
Cc: <stable@vger.kernel.org>, <stable-commits@vger.kernel.org>
Subject: Patch "[PATCH 4.4 12/16] cxlflash: Move to exponential back-off when cmd_room is not available" has been added to the 4.4-stable tree
Date: Fri, 09 Sep 2016 15:47:30 +0200	[thread overview]
Message-ID: <147342885025171@kroah.com> (raw)
In-Reply-To: <1472531709-18002-13-git-send-email-alexander.levin@verizon.com>


This is a note to let you know that I've just added the patch titled

    [PATCH 4.4 12/16] cxlflash: Move to exponential back-off when cmd_room is not available

to the 4.4-stable tree which can be found at:
    http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
     cxlflash-move-to-exponential-back-off-when-cmd_room-is-not-available.patch
and it can be found in the queue-4.4 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.


>From stable-owner@vger.kernel.org Tue Aug 30 06:36:51 2016
From: "Levin, Alexander" <alexander.levin@verizon.com>
Date: Tue, 30 Aug 2016 00:35:05 -0400
Subject: [PATCH 4.4 12/16] cxlflash: Move to exponential back-off when cmd_room is not available
To: "gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>
Cc: "stable@vger.kernel.org" <stable@vger.kernel.org>
Message-ID: <1472531709-18002-13-git-send-email-alexander.levin@verizon.com>


From: "Manoj N. Kumar" <manoj@linux.vnet.ibm.com>

[ Upstream commit ea76543127da32dec28af0a13ea1b06625fc085e ]

While profiling the cxlflash_queuecommand() path under a heavy load it
was found that number of retries to find cmd_room was fairly high.

There are two problems with the current back-off:
a) It starts with a udelay of 0
b) It backs-off linearly

Tried several approaches (a higher multiple 10*n, 100*n, as well as n^2,
2^n) and found that the exponential back-off(2^n) approach had the least
overall cost. Cost as being defined as overall time spent waiting.

The fix is to change the linear back-off to an exponential back-off.
This solution also takes care of the problem with the initial
delay (starts with 1 usec).

Signed-off-by: Manoj N. Kumar <manoj@linux.vnet.ibm.com>
Acked-by: Matthew R. Ochs <mrochs@linux.vnet.ibm.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Uma Krishnan <ukrishn@linux.vnet.ibm.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/scsi/cxlflash/main.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

--- a/drivers/scsi/cxlflash/main.c
+++ b/drivers/scsi/cxlflash/main.c
@@ -289,7 +289,7 @@ static void context_reset(struct afu_cmd
 		atomic64_set(&afu->room, room);
 		if (room)
 			goto write_rrin;
-		udelay(nretry);
+		udelay(1 << nretry);
 	} while (nretry++ < MC_ROOM_RETRY_CNT);
 
 	pr_err("%s: no cmd_room to send reset\n", __func__);
@@ -303,7 +303,7 @@ write_rrin:
 		if (rrin != 0x1)
 			break;
 		/* Double delay each time */
-		udelay(2 << nretry);
+		udelay(1 << nretry);
 	} while (nretry++ < MC_ROOM_RETRY_CNT);
 }
 
@@ -338,7 +338,7 @@ retry:
 			atomic64_set(&afu->room, room);
 			if (room)
 				goto write_ioarrin;
-			udelay(nretry);
+			udelay(1 << nretry);
 		} while (nretry++ < MC_ROOM_RETRY_CNT);
 
 		dev_err(dev, "%s: no cmd_room to send 0x%X\n",
@@ -352,7 +352,7 @@ retry:
 		 * afu->room.
 		 */
 		if (nretry++ < MC_ROOM_RETRY_CNT) {
-			udelay(nretry);
+			udelay(1 << nretry);
 			goto retry;
 		}
 


Patches currently in stable-queue which might be from alexander.levin@verizon.com are

queue-4.4/0065-megaraid_sas-Add-an-i-o-barrier.patch
queue-4.4/0110-ntp-Fix-ADJ_SETOFFSET-being-used-w-ADJ_NANO.patch
queue-4.4/0070-arm64-KVM-Configure-TCR_EL2.PS-at-runtime.patch
queue-4.4/0004-i40e-fix-do-not-sleep-in-netdev_ops.patch
queue-4.4/0125-tda10071-Fix-dependency-to-REGMAP_I2C.patch
queue-4.4/0009-i40e-clean-whole-mac-filter-list.patch
queue-4.4/cxlflash-fix-to-avoid-virtual-lun-failover-failure.patch
queue-4.4/0078-drm-atomic-Do-not-unset-crtc-when-an-encoder-is-stol.patch
queue-4.4/0017-clk-xgene-Fix-divider-with-non-zero-shift-value.patch
queue-4.4/0104-perf-x86-cqm-Fix-CQM-handling-of-grouping-events-int.patch
queue-4.4/0042-cxl-Fix-possible-idr-warning-when-contexts-are-relea.patch
queue-4.4/0131-drm-i915-Check-VBT-for-port-presence-in-addition-to-.patch
queue-4.4/0032-lightnvm-fix-missing-grown-bad-block-type.patch
queue-4.4/0076-drm-i915-bxt-update-list-of-PCIIDs.patch
queue-4.4/0071-net-cavium-liquidio-fix-check-for-in-progress-flag.patch
queue-4.4/0056-lpfc-Fix-mbox-reuse-in-PLOGI-completion.patch
queue-4.4/0090-UVC-Add-support-for-R200-depth-camera.patch
queue-4.4/0066-pwm-fsl-ftm-Fix-clock-enable-disable-when-using-PM.patch
queue-4.4/0013-igb-fix-NULL-derefs-due-to-skipped-SR-IOV-enabling.patch
queue-4.4/keys-fix-asn.1-indefinite-length-object-parsing.patch
queue-4.4/0107-net-thunderx-Fix-receive-packet-stats.patch
queue-4.4/0008-i40evf-check-rings-before-freeing-resources.patch
queue-4.4/0036-tools-hv-report-ENOSPC-errors-in-hv_fcopy_daemon.patch
queue-4.4/0093-net-mlx5e-Don-t-modify-CQ-before-it-was-created.patch
queue-4.4/0038-drivers-hv-cleanup-synic-msrs-if-vmbus-connect-faile.patch
queue-4.4/0083-net-thunderx-Fix-for-Qset-error-due-to-CQ-full.patch
queue-4.4/0133-netfilter-x_tables-check-for-size-overflow.patch
queue-4.4/0007-i40e-don-t-add-zero-MAC-filter.patch
queue-4.4/0030-lightnvm-unlock-rq-and-free-ppa_list-on-submission-f.patch
queue-4.4/0120-block-fix-blk_rq_get_max_sectors-for-driver-private-.patch
queue-4.4/0052-lpfc-Fix-driver-crash-when-module-parameter-lpfc_fcp.patch
queue-4.4/0034-Drivers-hv-util-Increase-the-timeout-for-util-servic.patch
queue-4.4/0119-irqchip-gicv3-its-numa-Enable-workaround-for-Cavium-.patch
queue-4.4/0010-i40e-Fix-Rx-hash-reported-to-the-stack-by-our-driver.patch
queue-4.4/0132-Bluetooth-Add-support-for-Intel-Bluetooth-device-826.patch
queue-4.4/0134-tipc-fix-an-infoleak-in-tipc_nl_compat_link_dump.patch
queue-4.4/0026-cxlflash-a-couple-off-by-one-bugs.patch
queue-4.4/0074-i40e-i40evf-Fix-RSS-rx-flow-hash-configuration-throu.patch
queue-4.4/0109-time-Verify-time-values-in-adjtimex-ADJ_SETOFFSET-to.patch
queue-4.4/0022-fm10k-always-check-init_hw-for-errors.patch
queue-4.4/0121-lpfc-Fix-DMA-faults-observed-upon-plugging-loopback-.patch
queue-4.4/0135-tipc-fix-nl-compat-regression-for-link-statistics.patch
queue-4.4/0049-lpfc-Fix-FCF-Infinite-loop-in-lpfc_sli4_fcf_rr_next_.patch
queue-4.4/0021-fm10k-reset-max_queues-on-init_hw_vf-failure.patch
queue-4.4/0067-pwm-lpc32xx-correct-number-of-PWM-channels-from-2-to.patch
queue-4.4/0002-i40evf-handle-many-MAC-filters-correctly.patch
queue-4.4/0055-lpfc-Fix-RDP-ACC-being-too-long.patch
queue-4.4/0128-net-thunderx-Fix-link-status-reporting.patch
queue-4.4/0029-lightnvm-add-check-after-mempool-allocation.patch
queue-4.4/0033-NVMe-fix-build-with-CONFIG_NVM-enabled.patch
queue-4.4/0031-lightnvm-fix-locking-and-mempool-in-rrpc_lun_gc.patch
queue-4.4/0023-fm10k-reinitialize-queuing-scheme-after-calling-init.patch
queue-4.4/0088-ALSA-hda-add-codec-support-for-Kabylake-display-audi.patch
queue-4.4/0115-cxlflash-Fix-to-resolve-dead-lock-during-EEH-recover.patch
queue-4.4/net-use-ns_capable_noaudit-when-determining-net-sysctl-permissions.patch
queue-4.4/0027-lightnvm-fix-bio-submission-issue.patch
queue-4.4/0096-arcmsr-fixes-not-release-allocated-resource.patch
queue-4.4/0045-cxlflash-Resolve-oops-in-wait_port_offline.patch
queue-4.4/0091-mmc-sdhci-Do-not-BUG-on-invalid-vdd.patch
queue-4.4/0097-Drivers-hv-vmbus-avoid-infinite-loop-in-init_vp_inde.patch
queue-4.4/0092-net-mlx5e-Don-t-try-to-modify-CQ-moderation-if-it-is.patch
queue-4.4/0101-intel_idle-Support-for-Intel-Xeon-Phi-Processor-x200.patch
queue-4.4/0035-Drivers-hv-utils-run-polling-callback-always-in-inte.patch
queue-4.4/0015-e1000-fix-data-race-between-tx_ring-next_to_clean.patch
queue-4.4/0014-ixgbe-Fix-handling-of-NAPI-budget-when-multiple-queu.patch
queue-4.4/0068-pwm-lpc32xx-fix-and-simplify-duty-cycle-and-period-c.patch
queue-4.4/0118-clocksource-Allow-unregistering-the-watchdog.patch
queue-4.4/crypto-nx-842-mask-xers0-bit-in-return-value.patch
queue-4.4/0124-crypto-vmx-Fix-ABI-detection.patch
queue-4.4/0051-lpfc-Fix-RegLogin-failed-error-seen-on-Lancer-FC-dur.patch
queue-4.4/0123-crypto-vmx-comply-with-ABIs-that-specify-vrsave-as-r.patch
queue-4.4/0025-fm10k-Cleanup-exception-handling-for-mailbox-interru.patch
queue-4.4/0054-lpfc-Fix-RDP-Speed-reporting.patch
queue-4.4/0057-lpfc-Fix-external-loopback-failure.patch
queue-4.4/0059-s390-cio-fix-measurement-characteristics-memleak.patch
queue-4.4/0061-s390-cio-update-measurement-characteristics.patch
queue-4.4/0094-s390-pci_dma-fix-DMA-table-corruption-with-4-TB-main.patch
queue-4.4/0073-mpt3sas-Fix-for-Asynchronous-completion-of-timedout-.patch
queue-4.4/0069-irqchip-gic-v3-Make-sure-read-from-ICC_IAR1_EL1-is-v.patch
queue-4.4/0130-drm-i915-Only-ignore-eDP-ports-that-are-connected.patch
queue-4.4/0113-ecryptfs-fix-handling-of-directory-opening.patch
queue-4.4/0063-megaraid_sas-Do-not-allow-PCI-access-during-OCR.patch
queue-4.4/0126-crypto-vmx-IV-size-failing-on-skcipher-API.patch
queue-4.4/0011-igb-don-t-unmap-NULL-hw_addr.patch
queue-4.4/0072-mpt3sas-A-correction-in-unmap_resources.patch
queue-4.4/0103-s390-crypto-provide-correct-file-mode-at-device-regi.patch
queue-4.4/0099-Drivers-hv-vmbus-don-t-manipulate-with-clocksources-.patch
queue-4.4/0111-drm-Balance-error-path-for-GEM-handle-allocation.patch
queue-4.4/0127-x86-hyperv-Avoid-reporting-bogus-NMI-status-for-Gen2.patch
queue-4.4/0058-qeth-initialize-net_device-with-carrier-off.patch
queue-4.4/0122-HID-core-prevent-out-of-bound-readings.patch
queue-4.4/0053-lpfc-Fix-crash-in-fcp-command-completion-path.patch
queue-4.4/netfilter-x_tables-check-for-size-overflow.patch
queue-4.4/cred-reject-inodes-with-invalid-ids-in-set_create_file_as.patch
queue-4.4/0077-drm-i915-skl-Add-missing-SKL-ids.patch
queue-4.4/0084-ahci-Workaround-for-ThunderX-Errata-22536.patch
queue-4.4/cxlflash-move-to-exponential-back-off-when-cmd_room-is-not-available.patch
queue-4.4/0018-fm10k-do-not-assume-VF-always-has-1-queue.patch
queue-4.4/0075-hrtimer-Catch-illegal-clockids.patch
queue-4.4/0080-qla2xxx-Use-ATIO-type-to-send-correct-tmr-response.patch
queue-4.4/ib-ipoib-do-not-set-skb-truesize-since-using-one-linearskb.patch
queue-4.4/kernel-add-noaudit-variant-of-ns_capable.patch
queue-4.4/0001-i40e-Workaround-fix-for-mss-256-issue.patch
queue-4.4/0006-i40e-properly-delete-VF-MAC-filters.patch
queue-4.4/0020-fm10k-Fix-handling-of-NAPI-budget-when-multiple-queu.patch
queue-4.4/0102-proc-revert-proc-pid-maps-stack-TID-annotation.patch
queue-4.4/0081-drm-amdgpu-fix-dp-link-rate-selection-v2.patch
queue-4.4/fs-check-for-invalid-i_uid-in-may_follow_link.patch
queue-4.4/0087-drm-i915-Fix-hpd-live-status-bits-for-g4x.patch
queue-4.4/0117-btrfs-Continue-write-in-case-of-can_not_nocow.patch
queue-4.4/0116-blk-mq-End-unstarted-requests-on-dying-queue.patch
queue-4.4/udp-properly-support-msg_peek-with-truncated-buffers.patch
queue-4.4/0085-arm64-Add-workaround-for-Cavium-erratum-27456.patch
queue-4.4/0089-sched-numa-Fix-use-after-free-bug-in-the-task_numa_c.patch
queue-4.4/0016-e1000e-fix-division-by-zero-on-jumbo-MTUs.patch
queue-4.4/0041-Drivers-hv-vmbus-fix-rescind-offer-handling-for-devi.patch
queue-4.4/0086-tipc-fix-nullptr-crash-during-subscription-cancel.patch
queue-4.4/0105-perf-x86-cqm-Fix-CQM-memory-leak-and-notifier-leak.patch
queue-4.4/0012-igb-use-the-correct-i210-register-for-EEMNGCTL.patch
queue-4.4/0098-Drivers-hv-vmbus-avoid-scheduling-in-interrupt-conte.patch
queue-4.4/0037-Drivers-hv-util-catch-allocation-errors.patch
queue-4.4/0039-Drivers-hv-vss-run-only-on-supported-host-versions.patch
queue-4.4/0064-megaraid_sas-Fix-SMAP-issue.patch
queue-4.4/0062-megaraid-Fix-possible-NULL-pointer-deference-in-mrai.patch
queue-4.4/0028-lightnvm-fix-incorrect-nr_free_blocks-stat.patch
queue-4.4/0048-cxl-Enable-PCI-device-ID-for-future-IBM-CXL-adapter.patch
queue-4.4/0095-arcmsr-fixed-getting-wrong-configuration-data.patch
queue-4.4/0114-drm-radeon-mst-fix-regression-in-lane-link-handling.patch
queue-4.4/0046-cxlflash-Enable-device-id-for-future-IBM-CXL-adapter.patch
queue-4.4/0112-ALSA-hda-add-AMD-Polaris-10-11-AZ-PCI-IDs-with-prope.patch
queue-4.4/0043-cxl-Fix-DSI-misses-when-the-context-owning-task-exit.patch
queue-4.4/0024-fm10k-Cleanup-MSI-X-interrupts-in-case-of-failure.patch
queue-4.4/0108-Input-xpad-correctly-handle-concurrent-LED-and-FF-re.patch
queue-4.4/0082-drm-radeon-fix-dp-link-rate-selection-v2.patch
queue-4.4/0047-cxl-fix-build-for-GCC-4.6.x.patch
queue-4.4/drivers-hv-lock-access-to-hyperv_mmio-resource-tree.patch
queue-4.4/0100-cxlflash-Fix-to-avoid-unnecessary-scan-with-internal.patch
queue-4.4/0019-fm10k-Correct-MTU-for-jumbo-frames.patch
queue-4.4/drm-amdgpu-cz-enable-disable-vce-dpm-even-if-vce-pg-is-disabled.patch
queue-4.4/0003-i40e-i40evf-Fix-RS-bit-update-in-Tx-path-and-disable.patch
queue-4.4/0050-lpfc-Fix-the-FLOGI-discovery-logic-to-comply-with-T1.patch
queue-4.4/cxlflash-fix-to-escalate-link_reset-also-on-port-1.patch
queue-4.4/0040-Drivers-hv-vmbus-serialize-process_chn_event-and-vmb.patch
queue-4.4/0005-i40e-Fix-memory-leaks-sideband-filter-programming.patch
queue-4.4/0106-net-thunderx-Fix-for-multiqset-not-configured-upon-i.patch
queue-4.4/0044-cxlflash-Fix-to-resolve-cmd-leak-after-host-reset.patch
queue-4.4/0079-mmc-sdhci-64-bit-DMA-actually-has-4-byte-alignment.patch
queue-4.4/0060-s390-cio-ensure-consistent-measurement-state.patch
queue-4.4/0129-Input-xpad-move-pending-clear-to-the-correct-locatio.patch

           reply	other threads:[~2016-09-09 13:47 UTC|newest]

Thread overview: expand[flat|nested]  mbox.gz  Atom feed
 [parent not found: <1472531709-18002-13-git-send-email-alexander.levin@verizon.com>]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=147342885025171@kroah.com \
    --to=gregkh@linuxfoundation.org \
    --cc=alexander.levin@verizon.com \
    --cc=jthumshirn@suse.de \
    --cc=manoj@linux.vnet.ibm.com \
    --cc=martin.petersen@oracle.com \
    --cc=mrochs@linux.vnet.ibm.com \
    --cc=stable-commits@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=ukrishn@linux.vnet.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.