[5.9,143/252] dmaengine: idxd: fix mapping of portal size

Message ID	20201123121842.499145755@linuxfoundation.org
State	New
Headers	show Return-Path: <stable-owner@kernel.org> From: Greg Kroah-Hartman <gregkh@linuxfoundation.org> To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>, stable@vger.kernel.org, Dave Jiang <dave.jiang@intel.com>, Vinod Koul <vkoul@kernel.org>, Sasha Levin <sashal@kernel.org> Subject: [PATCH 5.9 143/252] dmaengine: idxd: fix mapping of portal size Date: Mon, 23 Nov 2020 13:21:33 +0100 Message-Id: <20201123121842.499145755@linuxfoundation.org> In-Reply-To: <20201123121835.580259631@linuxfoundation.org> References: <20201123121835.580259631@linuxfoundation.org> User-Agent: quilt/0.66 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Precedence: bulk
Series	None \| expand [5.9,002/252] atm: nicstar: Unmap DMA on send error [5.9,003/252] bnxt_en: read EEPROM A2h address using page 0 [5.9,004/252] devlink: Add missing genlmsg_cancel() in devlink_nl_sb_port_pool_fill() [5.9,005/252] enetc: Workaround for MDIO register access issue [5.9,006/252] Exempt multicast addresses from five-second neighbor lifetime [5.9,007/252] inet_diag: Fix error path to cancel the meseage in inet_req_diag_fill() [5.9,008/252] ipv6: Fix error path to cancel the meseage [5.9,009/252] lan743x: fix issue causing intermittent kernel log warnings [5.9,010/252] lan743x: prevent entire kernel HANG on open, for some platforms [5.9,011/252] mlxsw: core: Use variable timeout for EMAD retries [5.9,012/252] net: b44: fix error return code in b44_init_one() [5.9,013/252] net: bridge: add missing counters to ndo_get_stats64 callback [5.9,014/252] netdevsim: set .owner to THIS_MODULE [5.9,015/252] net: dsa: mv88e6xxx: Avoid VTU corruption on 6097 [5.9,016/252] net: ethernet: mtk-star-emac: fix error return code in mtk_star_enable() [5.9,018/252] net: ethernet: ti: am65-cpts: update ret when ptp_clock is ERROR [5.9,019/252] net: ethernet: ti: cpsw: fix cpts irq after suspend [5.9,020/252] net: ethernet: ti: cpsw: fix error return code in cpsw_probe() [5.9,021/252] net: ftgmac100: Fix crash when removing driver [5.9,022/252] net: Have netpoll bring-up DSA management interface [5.9,023/252] net: ipa: lock when freeing transaction [5.9,024/252] netlabel: fix our progress tracking in netlbl_unlabel_staticlist() [5.9,025/252] netlabel: fix an uninitialized warning in netlbl_unlabel_staticlist() [5.9,026/252] net: lantiq: Wait for the GPHY firmware to be ready [5.9,027/252] net/mlx4_core: Fix init_hca fields offset [5.9,028/252] net/mlx5e: Fix refcount leak on kTLS RX resync [5.9,029/252] net/ncsi: Fix netlink registration [5.9,030/252] net: phy: mscc: remove non-MACSec compatible phy [5.9,031/252] net: qualcomm: rmnet: Fix incorrect receive packet handling during cleanup [5.9,032/252] net/smc: fix direct access to ib_gid_addr->ndev in smc_ib_determine_gid() [5.9,033/252] net: stmmac: Use rtnl_lock/unlock on netif_set_real_num_rx_queues() call [5.9,034/252] net/tls: fix corrupted data in recvmsg [5.9,035/252] net: x25: Increase refcnt of "struct x25_neigh" in x25_rx_call_request [5.9,036/252] page_frag: Recover from memory pressure [5.9,037/252] qed: fix error return code in qed_iwarp_ll2_start() [5.9,038/252] qed: fix ILT configuration of SRC block [5.9,039/252] qlcnic: fix error return code in qlcnic_83xx_restart_hw() [5.9,040/252] sctp: change to hold/put transport for proto_unreach_timer [5.9,041/252] tcp: only postpone PROBE_RTT if RTT is < current min_rtt estimate [5.9,042/252] vsock: forward all packets to the host when no H2G is registered [5.9,043/252] net/mlx5e: Fix check if netdev is bond slave [5.9,044/252] net/mlx5: Add handling of port type in rule deletion [5.9,045/252] net/mlx5: Clear bw_share upon VF disable [5.9,046/252] net/mlx5: Disable QoS when min_rates on all VFs are zero [5.9,047/252] PM: runtime: Add pm_runtime_resume_and_get to deal with usage counter [5.9,048/252] net: fec: Fix reference count leak in fec series ops [5.9,049/252] bnxt_en: Fix counter overflow logic. [5.9,050/252] bnxt_en: Free port stats during firmware reset. [5.9,051/252] net: mvneta: fix possible memory leak in mvneta_swbm_add_rx_fragment [5.9,052/252] net/tls: Fix wrong record sn in async mode of device resync [5.9,053/252] net: usb: qmi_wwan: Set DTR quirk for MR400 [5.9,054/252] Revert "Revert "gpio: omap: Fix lost edge wake-up interrupts"" [5.9,055/252] tools, bpftool: Avoid array index warnings. [5.9,056/252] habanalabs/gaudi: mask WDT error in QMAN [5.9,057/252] pinctrl: rockchip: enable gpio pclk for rockchip_gpio_to_irq [5.9,058/252] scsi: ufs: Fix unbalanced scsi_block_reqs_cnt caused by ufshcd_hold() [5.9,059/252] scsi: ufs: Try to save power mode change and UIC cmd completion timeout [5.9,060/252] pinctrl: mcp23s08: Print error message when regmap init fails [5.9,061/252] selftests: kvm: Fix the segment descriptor layout to match the actual layout [5.9,062/252] ACPI: button: Add DMI quirk for Medion Akoya E2228T [5.9,063/252] arm64: errata: Fix handling of 1418040 with late CPU onlining [5.9,064/252] arm64: psci: Avoid printing in cpu_psci_cpu_die() [5.9,065/252] arm64: smp: Tell RCU about CPUs that fail to come online [5.9,066/252] um: Call pgtable_pmd_page_dtor() in __pmd_free_tlb() [5.9,067/252] vfs: remove lockdep bogosity in __sb_start_write [5.9,068/252] gfs2: fix possible reference leak in gfs2_check_blk_type [5.9,069/252] hwmon: (pwm-fan) Fix RPM calculation [5.9,070/252] gfs2: Fix case in which ail writes are done to jdata holes [5.9,071/252] arm64: Add MIDR value for KRYO2XX gold/silver CPU cores [5.9,072/252] arm64: kpti: Add KRYO2XX gold/silver CPU cores to kpti safelist [5.9,073/252] arm64: cpu_errata: Apply Erratum 845719 to KRYO2XX Silver [5.9,074/252] usb: dwc2: Avoid leaving the error_debugfs label unused [5.9,075/252] arm64: dts: allwinner: beelink-gs1: Enable both RGMII RX/TX delay [5.9,076/252] arm64: dts: allwinner: Pine H64: Enable both RGMII RX/TX delay [5.9,077/252] arm64: dts: allwinner: a64: OrangePi Win: Fix ethernet node [5.9,078/252] arm64: dts: allwinner: a64: Pine64 Plus: Fix ethernet node [5.9,079/252] arm64: dts: allwinner: h5: OrangePi PC2: Fix ethernet node [5.9,080/252] ARM: dts: sun8i: r40: bananapi-m2-ultra: Fix ethernet node [5.9,081/252] Revert "arm: sun8i: orangepi-pc-plus: Set EMAC activity LEDs to active high" [5.9,082/252] ARM: dts: sun6i: a31-hummingbird: Enable RGMII RX/TX delay on Ethernet PHY [5.9,083/252] ARM: dts: sun7i: cubietruck: Enable RGMII RX/TX delay on Ethernet PHY [5.9,084/252] ARM: dts: sun7i: bananapi-m1-plus: Enable RGMII RX/TX delay on Ethernet PHY [5.9,085/252] ARM: dts: sun8i: h3: orangepi-plus2e: Enable RGMII RX/TX delay on Ethernet PHY [5.9,086/252] ARM: dts: sun8i: a83t: Enable both RGMII RX/TX delay on Ethernet PHY [5.9,087/252] ARM: dts: sun9i: Enable both RGMII RX/TX delay on Ethernet PHY [5.9,088/252] ARM: dts: sunxi: bananapi-m2-plus: Enable RGMII RX/TX delay on Ethernet PHY [5.9,089/252] arm64: dts: allwinner: h5: libretech-all-h5-cc: Enable RGMII RX/TX delay on PHY [5.9,090/252] arm64: dts: allwinner: a64: bananapi-m64: Enable RGMII RX/TX delay on PHY [5.9,091/252] Input: adxl34x - clean up a data type in adxl34x_probe() [5.9,092/252] MIPS: export has_transparent_hugepage() for modules [5.9,093/252] dmaengine: idxd: fix wq config registers offset programming [5.9,094/252] arm64: dts: allwinner: h5: OrangePi Prime: Fix ethernet node [5.9,095/252] arm64: dts: fsl: fix endianness issue of rcpm [5.9,096/252] arm64: dts: imx8mm-beacon-som: Fix Choppy BT audio [5.9,097/252] arm64: dts imx8mn: Remove non-existent USB OTG2 [5.9,098/252] arm: dts: imx6qdl-udoo: fix rgmii phy-mode for ksz9031 phy [5.9,099/252] ARM: dts: vf610-zii-dev-rev-b: Fix MDIO over clocking [5.9,100/252] ARM: dts: imx6q-prti6q: fix PHY address [5.9,101/252] swiotlb: using SIZE_MAX needs limits.h included [5.9,102/252] tee: amdtee: fix memory leak due to reset of global shm list [5.9,103/252] tee: amdtee: synchronize access to shm list [5.9,104/252] dmaengine: xilinx_dma: Fix usage of xilinx_aximcdma_tx_segment [5.9,105/252] dmaengine: xilinx_dma: Fix SG capability check for MCDMA [5.9,106/252] ARM: dts: stm32: Fix TA3-GPIO-C key on STM32MP1 DHCOM PDK2 [5.9,107/252] ARM: dts: stm32: Fix LED5 on STM32MP1 DHCOM PDK2 [5.9,108/252] ARM: dts: stm32: Define VIO regulator supply on DHCOM [5.9,109/252] ARM: dts: stm32: Enable thermal sensor support on stm32mp15xx-dhcor [5.9,110/252] ARM: dts: stm32: Keep VDDA LDO1 always on on DHCOM [5.9,111/252] arm64: dts: imx8mm: fix voltage for 1.6GHz CPU operating point [5.9,112/252] ARM: dts: imx50-evk: Fix the chip select 1 IOMUX [5.9,113/252] dmaengine: ti: omap-dma: Block PM if SDMA is busy to fix audio [5.9,114/252] kunit: tool: unmark test_data as binary blobs [5.9,115/252] rcu: Dont invoke try_invoke_on_locked_down_task() with irqs disabled [5.9,116/252] spi: fix client driver breakages when using GPIO descriptors [5.9,117/252] Input: resistive-adc-touch - fix kconfig dependency on IIO_BUFFER [5.9,118/252] Input: elan_i2c - fix firmware update on newer ICs [5.9,119/252] rfkill: Fix use-after-free in rfkill_resume() [5.9,120/252] RDMA/pvrdma: Fix missing kfree() in pvrdma_register_device() [5.9,121/252] RMDA/sw: Dont allow drivers using dma_virt_ops on highmem configs [5.9,122/252] perf lock: Correct field name "flags" [5.9,123/252] perf lock: Dont free "lock_seq_stat" if read_count isnt zero [5.9,124/252] SUNRPC: Fix oops in the rpc_xdr_buf event class [5.9,125/252] drm: bridge: dw-hdmi: Avoid resetting force in the detect function [5.9,126/252] tools, bpftool: Add missing close before bpftool net attach exit [5.9,127/252] IB/hfi1: Fix error return code in hfi1_init_dd() [5.9,128/252] ip_tunnels: Set tunnel option flag when tunnel metadata is present [5.9,129/252] can: af_can: prevent potential access of uninitialized member in can_rcv() [5.9,130/252] can: af_can: prevent potential access of uninitialized member in canfd_rcv() [5.9,131/252] can: dev: can_restart(): post buffer from the right context [5.9,132/252] can: ti_hecc: Fix memleak in ti_hecc_probe [5.9,133/252] can: mcba_usb: mcba_usb_start_xmit(): first fill skb, then pass to can_put_echo_skb() [5.9,134/252] can: peak_usb: fix potential integer overflow on shift of a int [5.9,135/252] can: flexcan: fix failure handling of pm_runtime_get_sync() [5.9,136/252] can: tcan4x5x: replace depends on REGMAP_SPI with depends on SPI [5.9,137/252] can: tcan4x5x: tcan4x5x_can_probe(): add missing error checking for devm_regmap_init() [5.9,138/252] can: tcan4x5x: tcan4x5x_can_remove(): fix order of deregistration [5.9,139/252] can: m_can: m_can_handle_state_change(): fix state change [5.9,140/252] can: m_can: m_can_class_free_dev(): introduce new function [5.9,141/252] can: m_can: Fix freeing of can device from peripherials [5.9,142/252] can: m_can: m_can_stop(): set device to software init mode before closing [5.9,143/252] dmaengine: idxd: fix mapping of portal size [5.9,144/252] ASoC: Intel: KMB: Fix S24_LE configuration [5.9,145/252] ASoC: qcom: lpass-platform: Fix memory leak [5.9,146/252] spi: cadence-quadspi: Fix error return code in cqspi_probe [5.9,147/252] selftests/bpf: Fix error return code in run_getsockopt_test() [5.9,148/252] MIPS: Alchemy: Fix memleak in alchemy_clk_setup_cpu [5.9,149/252] drm/sun4i: dw-hdmi: fix error return code in sun8i_dw_hdmi_bind() [5.9,150/252] net/mlx5: E-Switch, Fail mlx5_esw_modify_vport_rate if qos disabled [5.9,151/252] bpf, sockmap: Fix partial copy_page_to_iter so progress can still be made [5.9,152/252] bpf, sockmap: Ensure SO_RCVBUF memory is observed on ingress redirect [5.9,153/252] can: kvaser_pciefd: Fix KCAN bittiming limits [5.9,154/252] can: kvaser_usb: kvaser_usb_hydra: Fix KCAN bittiming limits [5.9,155/252] dmaengine: fix error codes in channel_register() [5.9,156/252] iommu/vt-d: Move intel_iommu_gfx_mapped to Intel IOMMU header [5.9,157/252] iommu/vt-d: Avoid panic if iommu init fails in tboot system [5.9,158/252] can: flexcan: flexcan_chip_start(): fix erroneous flexcan_transceiver_enable() duri... [5.9,159/252] can: m_can: process interrupt only when not runtime suspended [5.9,160/252] xfs: fix the minrecs logic when dealing with inode root child blocks [5.9,161/252] xfs: strengthen rmap record flags checking [5.9,162/252] xfs: directory scrub should check the null bestfree entries too [5.9,163/252] xfs: ensure inobt record walks always make forward progress [5.9,164/252] xfs: return corresponding errcode if xfs_initialize_perag() fail [5.9,165/252] ASOC: Intel: kbl_rt5663_rt5514_max98927: Do not try to disable disabled clock [5.9,166/252] regulator: ti-abb: Fix array out of bound read access on the first transition [5.9,167/252] libbpf: Fix VERSIONED_SYM_COUNT number parsing [5.9,168/252] lib/strncpy_from_user.c: Mask out bytes after NUL terminator. [5.9,169/252] fail_function: Remove a redundant mutex unlock [5.9,170/252] xfs: revert "xfs: fix rmap key and record comparison functions" [5.9,171/252] selftests/seccomp: powerpc: Fix typo in macro variable name [5.9,172/252] selftests/seccomp: sh: Fix register names [5.9,173/252] bpf, sockmap: Skb verdict SK_PASS to self already checked rmem limits [5.9,174/252] bpf, sockmap: On receive programs try to fast track SK_PASS ingress [5.9,175/252] bpf, sockmap: Use truesize with sk_rmem_schedule() [5.9,176/252] bpf, sockmap: Avoid returning unneeded EAGAIN when redirecting to self [5.9,177/252] efi/arm: set HSCTLR Thumb2 bit correctly for HVC calls from HYP [5.9,178/252] counter/ti-eqep: Fix regmap max_register [5.9,179/252] efi/x86: Free efi_pgd with free_pages() [5.9,180/252] sched/fair: Fix overutilized update in enqueue_task_fair() [5.9,181/252] sched: Fix data-race in wakeup [5.9,182/252] sched: Fix rq->nr_iowait ordering [5.9,183/252] libfs: fix error cast of negative value in simple_attr_write() [5.9,184/252] afs: Fix speculative status fetch going out of order wrt to modifications [5.9,185/252] HID: logitech-hidpp: Add PID for MX Anywhere 2 [5.9,186/252] HID: mcp2221: Fix GPIO output handling [5.9,187/252] HID: logitech-dj: Handle quad/bluetooth keyboards with a builtin trackpad [5.9,188/252] HID: logitech-dj: Fix Dinovo Mini when paired with a MX5x00 receiver [5.9,189/252] speakup: Do not let the line discipline be used several times [5.9,190/252] ALSA: firewire: Clean up a locking issue in copy_resp_to_buf() [5.9,191/252] ALSA: usb-audio: Add delay quirk for all Logitech USB devices [5.9,192/252] ALSA: ctl: fix error path at adding user-defined element set [5.9,193/252] ALSA: mixart: Fix mutex deadlock [5.9,194/252] ALSA: hda/realtek - Add supported for Lenovo ThinkPad Headset Button [5.9,195/252] ALSA: hda/realtek - Add supported mute Led for HP [5.9,196/252] ALSA: hda/realtek: Add some Clove SSID in the ALC293(ALC1220) [5.9,197/252] ALSA: hda/realtek - HP Headset Mic cant detect after boot [5.9,198/252] tty: serial: imx: fix potential deadlock [5.9,199/252] tty: serial: imx: keep console clocks always on [5.9,200/252] HID: logitech-dj: Fix an error in mse_bluetooth_descriptor [5.9,201/252] efivarfs: fix memory leak in efivarfs_create() [5.9,202/252] staging: rtl8723bs: Add 024c:0627 to the list of SDIO device-ids [5.9,203/252] staging: mt7621-pci: avoid to request pci bus resources [5.9,204/252] iio: light: fix kconfig dependency bug for VCNL4035 [5.9,205/252] ext4: fix bogus warning in ext4_update_dx_flag() [5.9,206/252] xfs: fix forkoff miscalculation related to XFS_LITINO(mp) [5.9,207/252] ACPI: fan: Initialize performance state sysfs attribute [5.9,208/252] iio: accel: kxcjk1013: Replace is_smo8500_device with an acpi_type enum [5.9,209/252] iio: accel: kxcjk1013: Add support for KIOX010A ACPI DSM for setting tablet-mode [5.9,210/252] iio: adc: mediatek: fix unset field [5.9,211/252] iio: cros_ec: Use default frequencies when EC returns invalid information [5.9,212/252] iio: imu: st_lsm6dsx: set 10ms as min shub slave timeout [5.9,213/252] iio/adc: ingenic: Fix AUX/VBAT readings when touchscreen is used [5.9,214/252] iio/adc: ingenic: Fix battery VREF for JZ4770 SoC [5.9,215/252] iio: adc: stm32-adc: fix a regression when using dma and irq [5.9,216/252] serial: ar933x_uart: disable clk on error handling path in probe [5.9,217/252] arm64: dts: agilex/stratix10: Fix qspi node compatible [5.9,218/252] spi: lpspi: Fix use-after-free on unbind [5.9,219/252] spi: Introduce device-managed SPI controller allocation [5.9,220/252] spi: npcm-fiu: Dont leak SPI master in probe error path [5.9,221/252] spi: bcm2835aux: Fix use-after-free on unbind [5.9,222/252] regulator: pfuze100: limit pfuze-support-disable-sw to pfuze{100,200} [5.9,223/252] regulator: fix memory leak with repeated set_machine_constraints() [5.9,224/252] regulator: avoid resolve_supply() infinite recursion [5.9,225/252] regulator: workaround self-referent regulators [5.9,226/252] gfs2: Fix regression in freeze_go_sync [5.9,227/252] xtensa: fix TLBTEMP area placement [5.9,228/252] xtensa: disable preemption around cache alias management calls [5.9,229/252] mac80211: minstrel: remove deferred sampling code [5.9,230/252] mac80211: minstrel: fix tx status processing corner case [5.9,231/252] mac80211: free sta in sta_info_insert_finish() on errors [5.9,232/252] s390: fix system call exit path [5.9,233/252] s390/cpum_sf.c: fix file permission for cpum_sfb_size [5.9,234/252] s390/dasd: fix null pointer dereference for ERP requests [5.9,235/252] Drivers: hv: vmbus: Allow cleanup of VMBUS_CONNECT_CPU if disconnected [5.9,236/252] drm/amd/display: Add missing pflip irq for dcn2.0 [5.9,237/252] drm/i915: Handle max_bpc==16 [5.9,238/252] drm/i915/tgl: Fix Media power gate sequence. [5.9,239/252] io_uring: dont double complete failed reissue request [5.9,240/252] mmc: sdhci-pci: Prefer SDR25 timing for High Speed mode for BYT-based Intel control... [5.9,241/252] mmc: sdhci-of-arasan: Allow configuring zero tap values [5.9,242/252] mmc: sdhci-of-arasan: Use Mask writes for Tap delays [5.9,243/252] mmc: sdhci-of-arasan: Issue DLL reset explicitly [5.9,244/252] blk-cgroup: fix a hd_struct leak in blkcg_fill_root_iostats [5.9,245/252] ptrace: Set PF_SUPERPRIV when checking capability [5.9,246/252] seccomp: Set PF_SUPERPRIV when checking capability [5.9,247/252] fanotify: fix logic of reporting name info with watched parent [5.9,248/252] x86/microcode/intel: Check patch signature before saving microcode for early loading [5.9,249/252] mm: never attempt async page lock if weve transferred data already [5.9,250/252] mm: fix readahead_page_batch for retry entries [5.9,251/252] mm: memcg/slab: fix root memcg vmstats [5.9,252/252] mm/userfaultfd: do not access vma->vm_mm after calling handle_userfault()

Message ID

20201123121842.499145755@linuxfoundation.org

State

New

Headers

From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Dave Jiang <dave.jiang@intel.com>,
	Vinod Koul <vkoul@kernel.org>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.9 143/252] dmaengine: idxd: fix mapping of portal size
Date: Mon, 23 Nov 2020 13:21:33 +0100
Message-Id: <20201123121842.499145755@linuxfoundation.org>
In-Reply-To: <20201123121835.580259631@linuxfoundation.org>
References: <20201123121835.580259631@linuxfoundation.org>
User-Agent: quilt/0.66
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

Series

None | expand

Commit Message

Greg KH Nov. 23, 2020, 12:21 p.m. UTC

From: Dave Jiang <dave.jiang@intel.com>

[ Upstream commit 8326be9f1c0bb498baf134878a8deb8a952e0135 ]

Portal size is 4k. Current code is mapping all 4 portals in a single chunk.
Restrict the mapped portal size to a single portal to ensure that submission
only goes to the intended portal address.

Fixes: c52ca478233c ("dmaengine: idxd: add configuration component of driver")
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/160513342642.510187.16450549281618747065.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/dma/idxd/device.c    | 2 +-
 drivers/dma/idxd/registers.h | 2 +-
 drivers/dma/idxd/submit.c    | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/dma/idxd/device.c b/drivers/dma/idxd/device.c
index c2beece445215..66e947627f569 100644
--- a/drivers/dma/idxd/device.c
+++ b/drivers/dma/idxd/device.c
@@ -271,7 +271,7 @@  int idxd_wq_map_portal(struct idxd_wq *wq)
 	resource_size_t start;
 
 	start = pci_resource_start(pdev, IDXD_WQ_BAR);
-	start = start + wq->id * IDXD_PORTAL_SIZE;
+	start += idxd_get_wq_portal_full_offset(wq->id, IDXD_PORTAL_LIMITED);
 
 	wq->dportal = devm_ioremap(dev, start, IDXD_PORTAL_SIZE);
 	if (!wq->dportal)
diff --git a/drivers/dma/idxd/registers.h b/drivers/dma/idxd/registers.h
index aef5a902829ee..54390334c243a 100644
--- a/drivers/dma/idxd/registers.h
+++ b/drivers/dma/idxd/registers.h
@@ -8,7 +8,7 @@ 
 
 #define IDXD_MMIO_BAR		0
 #define IDXD_WQ_BAR		2
-#define IDXD_PORTAL_SIZE	0x4000
+#define IDXD_PORTAL_SIZE	PAGE_SIZE
 
 /* MMIO Device BAR0 Registers */
 #define IDXD_VER_OFFSET			0x00
diff --git a/drivers/dma/idxd/submit.c b/drivers/dma/idxd/submit.c
index 156a1ee233aa5..417048e3c42aa 100644
--- a/drivers/dma/idxd/submit.c
+++ b/drivers/dma/idxd/submit.c
@@ -74,7 +74,7 @@  int idxd_submit_desc(struct idxd_wq *wq, struct idxd_desc *desc)
 	if (idxd->state != IDXD_DEV_ENABLED)
 		return -EIO;
 
-	portal = wq->dportal + idxd_get_wq_portal_offset(IDXD_PORTAL_UNLIMITED);
+	portal = wq->dportal;
 	/*
 	 * The wmb() flushes writes to coherent DMA data before possibly
 	 * triggering a DMA read. The wmb() is necessary even on UP because

[5.9,143/252] dmaengine: idxd: fix mapping of portal size

Commit Message

Patch