June 2020 - Linux-stable-mirror

[PATCH] media: cedrus: Propagate OUTPUT resolution to CAPTURE

by Nicolas Dufresne

As per spec, the CAPTURE resolution should be automatically set based on the OTUPUT resolution. This patch properly propagate width/height to the capture when the OUTPUT format is set and override the user provided width/height with configured OUTPUT resolution when the CAPTURE fmt is updated. This also prevents userspace from selecting a CAPTURE resolution that is too small, avoiding unwanted page faults. Signed-off-by: Nicolas Dufresne <nicolas.dufresne(a)collabora.com> --- drivers/staging/media/sunxi/cedrus/cedrus_video.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/staging/media/sunxi/cedrus/cedrus_video.c b/drivers/staging/media/sunxi/cedrus/cedrus_video.c index 16d82309e7b6..a6d6b15adc2e 100644 --- a/drivers/staging/media/sunxi/cedrus/cedrus_video.c +++ b/drivers/staging/media/sunxi/cedrus/cedrus_video.c @@ -247,6 +247,8 @@ static int cedrus_try_fmt_vid_cap(struct file *file, void *priv, return -EINVAL; pix_fmt->pixelformat = fmt->pixelformat; + pix_fmt->width = ctx->src_fmt.width; + pix_fmt->height = ctx->src_fmt.height; cedrus_prepare_format(pix_fmt); return 0; @@ -319,11 +321,14 @@ static int cedrus_s_fmt_vid_out(struct file *file, void *priv, break; } - /* Propagate colorspace information to capture. */ + /* Propagate format information to capture. */ ctx->dst_fmt.colorspace = f->fmt.pix.colorspace; ctx->dst_fmt.xfer_func = f->fmt.pix.xfer_func; ctx->dst_fmt.ycbcr_enc = f->fmt.pix.ycbcr_enc; ctx->dst_fmt.quantization = f->fmt.pix.quantization; + ctx->dst_fmt.width = ctx->src_fmt.width; + ctx->dst_fmt.height = ctx->src_fmt.height; + cedrus_prepare_format(&ctx->dst_fmt); return 0; } -- 2.26.2

4 years, 5 months

3
3
0 0

[PATCH 4/4] ovl: inode reference leak in ovl_is_inuse true case.

by youngjun

When "ovl_is_inuse" true case, trap inode reference not put. plus adding the comment explaining sequence of ovl_is_inuse after ovl_setup_trap. Fixes: 0be0bfd2de9d ("ovl: fix regression caused by overlapping layers..") Cc: <stable(a)vger.kernel.org> # v4.19+ Reviewed-by: Amir Goldstein <amir73il(a)gmail.com> Signed-off-by: youngjun <her0gyugyu(a)gmail.com> --- fs/overlayfs/super.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c index 91476bc422f9..3097142b1e23 100644 --- a/fs/overlayfs/super.c +++ b/fs/overlayfs/super.c @@ -1493,14 +1493,22 @@ static int ovl_get_layers(struct super_block *sb, struct ovl_fs *ofs, if (err < 0) goto out; + /* + * Check if lower root conflicts with this overlay layers before checking + * if it is in-use as upperdir/workdir of "another" mount, because we do + * not bother to check in ovl_is_inuse() if the upperdir/workdir is in fact + * in-use by our upperdir/workdir. + */ err = ovl_setup_trap(sb, stack[i].dentry, &trap, "lowerdir"); if (err) goto out; if (ovl_is_inuse(stack[i].dentry)) { err = ovl_report_in_use(ofs, "lowerdir"); - if (err) + if (err) { + iput(trap); goto out; + } } mnt = clone_private_mount(&stack[i]); -- 2.17.1 Again, Great thanks Amir. I revise my patch through your kind guidance.

4 years, 5 months

2
1
0 0

[PATCH 4.19] IB/umem: fix reference count leak in ib_umem_odp_get()

by Yang Yingliang

Add missing mmput() on error path to avoid ref-count leak. This problem has already been resolved in mainline by f27a0d50a4bc ("RDMA/umem: Use umem->owning_mm inside ODP"). Fixes: 79bb5b7ee177 ("RDMA/umem: Fix missing mmap_sem in get umem ODP call") Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/infiniband/core/umem_odp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c index eeafdc0beec7..08ef654ea9b8 100644 --- a/drivers/infiniband/core/umem_odp.c +++ b/drivers/infiniband/core/umem_odp.c @@ -347,7 +347,8 @@ int ib_umem_odp_get(struct ib_ucontext *context, struct ib_umem *umem, vma = find_vma(mm, ib_umem_start(umem)); if (!vma || !is_vm_hugetlb_page(vma)) { up_read(&mm->mmap_sem); - return -EINVAL; + ret_val = -EINVAL; + goto out_mm; } h = hstate_vma(vma); umem->page_shift = huge_page_shift(h); -- 2.25.1

4 years, 5 months

2
2
0 0

[PATCH v3] ARM: dts: imx6qdl-gw551x: fix audio SSI

by Tim Harvey

The audio codec on the GW551x routes to ssi1 Cc: stable(a)vger.kernel.org Fixes: 3117e851cef1 ("ARM: dts: imx: Add TDA19971 HDMI Receiver to GW551x") Signed-off-by: Tim Harvey <tharvey(a)gateworks.com> --- v3: add fixes tag v2: fix typo in commit short desc --- arch/arm/boot/dts/imx6qdl-gw551x.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/boot/dts/imx6qdl-gw551x.dtsi b/arch/arm/boot/dts/imx6qdl-gw551x.dtsi index c38e86e..8c33510 100644 --- a/arch/arm/boot/dts/imx6qdl-gw551x.dtsi +++ b/arch/arm/boot/dts/imx6qdl-gw551x.dtsi @@ -110,7 +110,7 @@ simple-audio-card,frame-master = <&sound_codec>; sound_cpu: simple-audio-card,cpu { - sound-dai = <&ssi2>; + sound-dai = <&ssi1>; }; sound_codec: simple-audio-card,codec { -- 2.7.4

4 years, 5 months

3
2
0 0

[PATCH AUTOSEL 4.19 001/106] ath9x: Fix stack-out-of-bounds Write in ath9k_hif_usb_rx_cb

by Sasha Levin

From: Qiujun Huang <hqjagain(a)gmail.com> [ Upstream commit 19d6c375d671ce9949a864fb9a03e19f5487b4d3 ] Add barrier to accessing the stack array skb_pool. The case reported by syzbot: https://lore.kernel.org/linux-usb/0000000000003d7c1505a2168418@google.com BUG: KASAN: stack-out-of-bounds in ath9k_hif_usb_rx_stream drivers/net/wireless/ath/ath9k/hif_usb.c:626 [inline] BUG: KASAN: stack-out-of-bounds in ath9k_hif_usb_rx_cb+0xdf6/0xf70 drivers/net/wireless/ath/ath9k/hif_usb.c:666 Write of size 8 at addr ffff8881db309a28 by task swapper/1/0 Call Trace: ath9k_hif_usb_rx_stream drivers/net/wireless/ath/ath9k/hif_usb.c:626 [inline] ath9k_hif_usb_rx_cb+0xdf6/0xf70 drivers/net/wireless/ath/ath9k/hif_usb.c:666 __usb_hcd_giveback_urb+0x1f2/0x470 drivers/usb/core/hcd.c:1648 usb_hcd_giveback_urb+0x368/0x420 drivers/usb/core/hcd.c:1713 dummy_timer+0x1258/0x32ae drivers/usb/gadget/udc/dummy_hcd.c:1966 call_timer_fn+0x195/0x6f0 kernel/time/timer.c:1404 expire_timers kernel/time/timer.c:1449 [inline] __run_timers kernel/time/timer.c:1773 [inline] __run_timers kernel/time/timer.c:1740 [inline] run_timer_softirq+0x5f9/0x1500 kernel/time/timer.c:1786 Reported-and-tested-by: syzbot+d403396d4df67ad0bd5f(a)syzkaller.appspotmail.com Signed-off-by: Qiujun Huang <hqjagain(a)gmail.com> Signed-off-by: Kalle Valo <kvalo(a)codeaurora.org> Link: https://lore.kernel.org/r/20200404041838.10426-5-hqjagain@gmail.com Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- drivers/net/wireless/ath/ath9k/hif_usb.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.c b/drivers/net/wireless/ath/ath9k/hif_usb.c index dd0c32379375..c4a2b7201ce3 100644 --- a/drivers/net/wireless/ath/ath9k/hif_usb.c +++ b/drivers/net/wireless/ath/ath9k/hif_usb.c @@ -612,6 +612,11 @@ static void ath9k_hif_usb_rx_stream(struct hif_device_usb *hif_dev, hif_dev->remain_skb = nskb; spin_unlock(&hif_dev->rx_lock); } else { + if (pool_index == MAX_PKT_NUM_IN_TRANSFER) { + dev_err(&hif_dev->udev->dev, + "ath9k_htc: over RX MAX_PKT_NUM\n"); + goto err; + } nskb = __dev_alloc_skb(pkt_len + 32, GFP_ATOMIC); if (!nskb) { dev_err(&hif_dev->udev->dev, -- 2.25.1

4 years, 5 months

2
106
0 0

[PATCH 1/3] arm64: dts: agilex: add status to qspi dts node

by Dinh Nguyen

Add status = "okay" to QSPI node. Fixes: c4c8757b2d895 ("arm64: dts: agilex: add QSPI support for Intel Agilex") Cc: linux-stable <stable(a)vger.kernel.org> # >= v5.5 Signed-off-by: Dinh Nguyen <dinguyen(a)kernel.org> --- arch/arm64/boot/dts/intel/socfpga_agilex_socdk.dts | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm64/boot/dts/intel/socfpga_agilex_socdk.dts b/arch/arm64/boot/dts/intel/socfpga_agilex_socdk.dts index 51d948323bfd..92f478def723 100644 --- a/arch/arm64/boot/dts/intel/socfpga_agilex_socdk.dts +++ b/arch/arm64/boot/dts/intel/socfpga_agilex_socdk.dts @@ -98,6 +98,7 @@ }; &qspi { + status = "okay"; flash@0 { #address-cells = <1>; #size-cells = <1>; -- 2.17.1

4 years, 5 months

2
5
0 0

[PATCH] omapfb: dss: Fix max fclk divider for omap36xx

by Adam Ford

The drm/omap driver was fixed to correct an issue where using a divider of 32 breaks the DSS despite the TRM stating 32 is a valid number. Through experimentation, it appears that 31 works, and it is consistent with the value used by the drm/omap driver. This patch fixes the divider for fbdev driver instead of the drm. Fixes: f76ee892a99e ("omapfb: copy omapdss & displays for omapfb") Cc: <stable(a)vger.kernel.org> #4.9+ Signed-off-by: Adam Ford <aford173(a)gmail.com> --- Linux 4.4 will need a similar patch, but it doesn't apply cleanly. The DRM version of this same fix is: e2c4ed148cf3 ("drm/omap: fix max fclk divider for omap36xx") diff --git a/drivers/video/fbdev/omap2/omapfb/dss/dss.c b/drivers/video/fbdev/omap2/omapfb/dss/dss.c index 7252d22dd117..bfc5c4c5a26a 100644 --- a/drivers/video/fbdev/omap2/omapfb/dss/dss.c +++ b/drivers/video/fbdev/omap2/omapfb/dss/dss.c @@ -833,7 +833,7 @@ static const struct dss_features omap34xx_dss_feats = { }; static const struct dss_features omap3630_dss_feats = { - .fck_div_max = 32, + .fck_div_max = 31, .dss_fck_multiplier = 1, .parent_clk_name = "dpll4_ck", .dpi_select_source = &dss_dpi_select_source_omap2_omap3, -- 2.25.1

4 years, 5 months

5
7
0 0

[PATCH 04/10] mfd: altera-sysmgr: Fix physical address storing hacks

by Lee Jones

Sparse reports: drivers/mfd/altera-sysmgr.c:150:30: warning: incorrect type in assignment (different address spaces) drivers/mfd/altera-sysmgr.c:150:30: expected unsigned int [usertype] *base drivers/mfd/altera-sysmgr.c:150:30: got void [noderef] <asn:2> * drivers/mfd/altera-sysmgr.c:156:26: warning: incorrect type in argument 3 (different address spaces) drivers/mfd/altera-sysmgr.c:156:26: expected void [noderef] <asn:2> *regs drivers/mfd/altera-sysmgr.c:156:26: got unsigned int [usertype] *base It appears as though the driver data property 'resource_size_t *base' was being used to store 2 different types of addresses (physical and IO-mapped) under a single declared type. Fortunately, no value is recalled from the driver data entry, so it can be easily omitted. Instead we can use the value obtained directly from the platform resource to pass through Regmap into the call-backs to be used for the SMCC call and use a local dedicated __iomem variable for IO-remapping. Cc: <stable(a)vger.kernel.org> Cc: Thor Thayer <thor.thayer(a)linux.intel.com> Signed-off-by: Lee Jones <lee.jones(a)linaro.org> --- drivers/mfd/altera-sysmgr.c | 16 ++++++---------- 1 file changed, 6 insertions(+), 10 deletions(-) diff --git a/drivers/mfd/altera-sysmgr.c b/drivers/mfd/altera-sysmgr.c index d2a13a547a3ca..83f0765f819b0 100644 --- a/drivers/mfd/altera-sysmgr.c +++ b/drivers/mfd/altera-sysmgr.c @@ -22,11 +22,9 @@ /** * struct altr_sysmgr - Altera SOCFPGA System Manager * @regmap: the regmap used for System Manager accesses. - * @base : the base address for the System Manager */ struct altr_sysmgr { struct regmap *regmap; - resource_size_t *base; }; static struct platform_driver altr_sysmgr_driver; @@ -127,6 +125,7 @@ static int sysmgr_probe(struct platform_device *pdev) struct regmap_config sysmgr_config = altr_sysmgr_regmap_cfg; struct device *dev = &pdev->dev; struct device_node *np = dev->of_node; + void __iomem *base; sysmgr = devm_kzalloc(dev, sizeof(*sysmgr), GFP_KERNEL); if (!sysmgr) @@ -139,22 +138,19 @@ static int sysmgr_probe(struct platform_device *pdev) sysmgr_config.max_register = resource_size(res) - sysmgr_config.reg_stride; if (of_device_is_compatible(np, "altr,sys-mgr-s10")) { - /* Need physical address for SMCC call */ - sysmgr->base = (resource_size_t *)res->start; sysmgr_config.reg_read = s10_protected_reg_read; sysmgr_config.reg_write = s10_protected_reg_write; - regmap = devm_regmap_init(dev, NULL, sysmgr->base, + /* Need physical address for SMCC call */ + regmap = devm_regmap_init(dev, NULL, (void *)res->start, &sysmgr_config); } else { - sysmgr->base = devm_ioremap(dev, res->start, - resource_size(res)); - if (!sysmgr->base) + base = devm_ioremap(dev, res->start, resource_size(res)); + if (!base) return -ENOMEM; sysmgr_config.max_register = res->end - res->start - 3; - regmap = devm_regmap_init_mmio(dev, sysmgr->base, - &sysmgr_config); + regmap = devm_regmap_init_mmio(dev, base, &sysmgr_config); } if (IS_ERR(regmap)) { -- 2.25.1

4 years, 5 months

2
3
0 0

[PATCH for-rc v2 1/2] IB/hfi1: Do not destroy hfi1_wq when the device is shut down

by Dennis Dalessandro

From: Kaike Wan <kaike.wan(a)intel.com> The workqueue hfi1_wq is destroyed in function shutdown_device(), which is called by either shutdown_one() or remove_one(). The function shutdown_one() is called when the kernel is rebooted while remove_one() is called when the hfi1 driver is unloaded. When the kernel is rebooted, hfi1_wq is destroyed while all qps are still active, leading to a kernel crash: [ 198.891986] BUG: unable to handle kernel NULL pointer dereference at 0000000000000102 [ 198.892072] IP: [<ffffffff94cb7b02>] __queue_work+0x32/0x3e0 [ 198.892130] PGD 0 [ 198.892156] Oops: 0000 [#1] SMP [ 198.892193] Modules linked in: dm_round_robin nvme_rdma(OE) nvme_fabrics(OE) nvme_core(OE) ib_isert iscsi_target_mod target_core_mod ib_ucm mlx4_ib iTCO_wdt iTCO_vendor_support mxm_wmi sb_edac intel_powerclamp coretemp intel_rapl iosf_mbi kvm rpcrdma sunrpc irqbypass crc32_pclmul ghash_clmulni_intel rdma_ucm aesni_intel ib_uverbs lrw gf128mul opa_vnic glue_helper ablk_helper ib_iser cryptd ib_umad rdma_cm iw_cm ses enclosure libiscsi scsi_transport_sas pcspkr joydev ib_ipoib(OE) scsi_transport_iscsi ib_cm sg ipmi_ssif mei_me lpc_ich i2c_i801 mei ioatdma ipmi_si dm_multipath ipmi_devintf ipmi_msghandler wmi acpi_pad acpi_power_meter hangcheck_timer ip_tables ext4 mbcache jbd2 mlx4_en sd_mod crc_t10dif crct10dif_generic mgag200 drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm hfi1(OE) [ 198.892997] crct10dif_pclmul crct10dif_common crc32c_intel drm ahci mlx4_core libahci rdmavt(OE) igb megaraid_sas ib_core libata drm_panel_orientation_quirks ptp pps_core devlink dca i2c_algo_bit dm_mirror dm_region_hash dm_log dm_mod [ 198.893233] CPU: 19 PID: 0 Comm: swapper/19 Kdump: loaded Tainted: G OE ------------ 3.10.0-957.el7.x86_64 #1 [ 198.893317] Hardware name: Phegda X2226A/S2600CW, BIOS SE5C610.86B.01.01.0024.021320181901 02/13/2018 [ 198.893388] task: ffff8a799ba0d140 ti: ffff8a799bad8000 task.ti: ffff8a799bad8000 [ 198.893447] RIP: 0010:[<ffffffff94cb7b02>] [<ffffffff94cb7b02>] __queue_work+0x32/0x3e0 [ 198.893518] RSP: 0018:ffff8a90dde43d80 EFLAGS: 00010046 [ 198.893561] RAX: 0000000000000082 RBX: 0000000000000086 RCX: 0000000000000000 [ 198.893617] RDX: ffff8a90b924fcb8 RSI: 0000000000000000 RDI: 000000000000001b [ 198.893674] RBP: ffff8a90dde43db8 R08: ffff8a799ba0d6d8 R09: ffff8a90dde53900 [ 198.893730] R10: 0000000000000002 R11: ffff8a90dde43de8 R12: ffff8a90b924fcb8 [ 198.893786] R13: 000000000000001b R14: 0000000000000000 R15: ffff8a90d2890000 [ 198.893843] FS: 0000000000000000(0000) GS:ffff8a90dde40000(0000) knlGS:0000000000000000 [ 198.893905] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 198.893953] CR2: 0000000000000102 CR3: 0000001a70410000 CR4: 00000000001607e0 [ 198.894009] Call Trace: [ 198.894058] [<ffffffff94cb8105>] queue_work_on+0x45/0x50 [ 198.894161] [<ffffffffc03f781e>] _hfi1_schedule_send+0x6e/0xc0 [hfi1] [ 198.894244] [<ffffffffc03f78a2>] hfi1_schedule_send+0x32/0x70 [hfi1] [ 198.894311] [<ffffffffc02cf2d9>] rvt_rc_timeout+0xe9/0x130 [rdmavt] [ 198.894366] [<ffffffff94ce563a>] ? trigger_load_balance+0x6a/0x280 [ 198.894426] [<ffffffffc02cf1f0>] ? rvt_free_qpn+0x40/0x40 [rdmavt] [ 198.894482] [<ffffffff94ca7f58>] call_timer_fn+0x38/0x110 [ 198.894537] [<ffffffffc02cf1f0>] ? rvt_free_qpn+0x40/0x40 [rdmavt] [ 198.894589] [<ffffffff94caa3bd>] run_timer_softirq+0x24d/0x300 [ 198.894641] [<ffffffff94ca0f05>] __do_softirq+0xf5/0x280 [ 198.894689] [<ffffffff9537832c>] call_softirq+0x1c/0x30 [ 198.894736] [<ffffffff94c2e675>] do_softirq+0x65/0xa0 [ 198.894780] [<ffffffff94ca1285>] irq_exit+0x105/0x110 [ 198.894826] [<ffffffff953796c8>] smp_apic_timer_interrupt+0x48/0x60 [ 198.894881] [<ffffffff95375df2>] apic_timer_interrupt+0x162/0x170 [ 198.894930] <EOI> [ 198.894955] [<ffffffff951adfb7>] ? cpuidle_enter_state+0x57/0xd0 [ 198.895012] [<ffffffff951ae10e>] cpuidle_idle_call+0xde/0x230 [ 198.895063] [<ffffffff94c366de>] arch_cpu_idle+0xe/0xc0 [ 198.896984] [<ffffffff94cfc3ba>] cpu_startup_entry+0x14a/0x1e0 [ 198.898879] [<ffffffff94c57db7>] start_secondary+0x1f7/0x270 [ 198.900750] [<ffffffff94c000d5>] start_cpu+0x5/0x14 The solution is to destroy the workqueue only when the hfi1 driver is unloaded, not when the device is shut down. In addition, when the device is shut down, no more work should be scheduled on the workqueues and the workqueues are flushed. Fixes: 8d3e71136a08 ("IB/{hfi1, qib}: Add handling of kernel restart") Cc: <stable(a)vger.kernel.org> Reviewed-by: Mike Marciniszyn <mike.marciniszyn(a)intel.com> Signed-off-by: Kaike Wan <kaike.wan(a)intel.com> Signed-off-by: Dennis Dalessandro <dennis.dalessandro(a)intel.com> --- Changes since v1: - Flush the workqueues when the device is shut down; - Stop queuing more work on hfi1_wq when the device is shut down. --- drivers/infiniband/hw/hfi1/init.c | 27 +++++++++++++++++++++++---- drivers/infiniband/hw/hfi1/qp.c | 5 ++++- drivers/infiniband/hw/hfi1/tid_rdma.c | 5 ++++- 3 files changed, 31 insertions(+), 6 deletions(-) diff --git a/drivers/infiniband/hw/hfi1/init.c b/drivers/infiniband/hw/hfi1/init.c index 5eed436..16d6788 100644 --- a/drivers/infiniband/hw/hfi1/init.c +++ b/drivers/infiniband/hw/hfi1/init.c @@ -831,6 +831,25 @@ static int create_workqueues(struct hfi1_devdata *dd) } /** + * destroy_workqueues - destroy per port workqueues + * @dd: the hfi1_ib device + */ +static void destroy_workqueues(struct hfi1_devdata *dd) +{ + int pidx; + struct hfi1_pportdata *ppd; + + for (pidx = 0; pidx < dd->num_pports; ++pidx) { + ppd = dd->pport + pidx; + + if (ppd->hfi1_wq) { + destroy_workqueue(ppd->hfi1_wq); + ppd->hfi1_wq = NULL; + } + } +} + +/** * enable_general_intr() - Enable the IRQs that will be handled by the * general interrupt handler. * @dd: valid devdata @@ -1104,11 +1123,10 @@ static void shutdown_device(struct hfi1_devdata *dd) */ hfi1_quiet_serdes(ppd); - if (ppd->hfi1_wq) { - destroy_workqueue(ppd->hfi1_wq); - ppd->hfi1_wq = NULL; - } + if (ppd->hfi1_wq) + flush_workqueue(ppd->hfi1_wq); if (ppd->link_wq) { + flush_workqueue(ppd->link_wq); destroy_workqueue(ppd->link_wq); ppd->link_wq = NULL; } @@ -1756,6 +1774,7 @@ static void remove_one(struct pci_dev *pdev) * clear dma engines, etc. */ shutdown_device(dd); + destroy_workqueues(dd); stop_timers(dd); diff --git a/drivers/infiniband/hw/hfi1/qp.c b/drivers/infiniband/hw/hfi1/qp.c index 0c2ae9f..283d4c5 100644 --- a/drivers/infiniband/hw/hfi1/qp.c +++ b/drivers/infiniband/hw/hfi1/qp.c @@ -367,7 +367,10 @@ bool _hfi1_schedule_send(struct rvt_qp *qp) struct hfi1_ibport *ibp = to_iport(qp->ibqp.device, qp->port_num); struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); - struct hfi1_devdata *dd = dd_from_ibdev(qp->ibqp.device); + struct hfi1_devdata *dd = ppd->dd; + + if (dd->flags & HFI1_SHUTDOWN) + return true; return iowait_schedule(&priv->s_iowait, ppd->hfi1_wq, priv->s_sde ? diff --git a/drivers/infiniband/hw/hfi1/tid_rdma.c b/drivers/infiniband/hw/hfi1/tid_rdma.c index 243b4ba..facff13 100644 --- a/drivers/infiniband/hw/hfi1/tid_rdma.c +++ b/drivers/infiniband/hw/hfi1/tid_rdma.c @@ -5406,7 +5406,10 @@ static bool _hfi1_schedule_tid_send(struct rvt_qp *qp) struct hfi1_ibport *ibp = to_iport(qp->ibqp.device, qp->port_num); struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); - struct hfi1_devdata *dd = dd_from_ibdev(qp->ibqp.device); + struct hfi1_devdata *dd = ppd->dd; + + if ((dd->flags & HFI1_SHUTDOWN)) + return true; return iowait_tid_schedule(&priv->s_iowait, ppd->hfi1_wq, priv->s_sde ?

4 years, 5 months

2
2
0 0

[PATCH v3 1/3] mm/shuffle: don't move pages between zones and don't read garbage memmaps

by David Hildenbrand

Especially with memory hotplug, we can have offline sections (with a garbage memmap) and overlapping zones. We have to make sure to only touch initialized memmaps (online sections managed by the buddy) and that the zone matches, to not move pages between zones. To test if this can actually happen, I added a simple BUG_ON(page_zone(page_i) != page_zone(page_j)); right before the swap. When hotplugging a 256M DIMM to a 4G x86-64 VM and onlining the first memory block "online_movable" and the second memory block "online_kernel", it will trigger the BUG, as both zones (NORMAL and MOVABLE) overlap. This might result in all kinds of weird situations (e.g., double allocations, list corruptions, unmovable allocations ending up in the movable zone). Fixes: e900a918b098 ("mm: shuffle initial free memory to improve memory-side-cache utilization") Reviewed-by: Wei Yang <richard.weiyang(a)linux.alibaba.com> Acked-by: Michal Hocko <mhocko(a)suse.com> Acked-by: Dan Williams <dan.j.williams(a)intel.com> Cc: stable(a)vger.kernel.org # v5.2+ Cc: Andrew Morton <akpm(a)linux-foundation.org> Cc: Johannes Weiner <hannes(a)cmpxchg.org> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Minchan Kim <minchan(a)kernel.org> Cc: Huang Ying <ying.huang(a)intel.com> Cc: Wei Yang <richard.weiyang(a)gmail.com> Cc: Mel Gorman <mgorman(a)techsingularity.net> Signed-off-by: David Hildenbrand <david(a)redhat.com> --- mm/shuffle.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/mm/shuffle.c b/mm/shuffle.c index 44406d9977c77..dd13ab851b3ee 100644 --- a/mm/shuffle.c +++ b/mm/shuffle.c @@ -58,25 +58,25 @@ module_param_call(shuffle, shuffle_store, shuffle_show, &shuffle_param, 0400); * For two pages to be swapped in the shuffle, they must be free (on a * 'free_area' lru), have the same order, and have the same migratetype. */ -static struct page * __meminit shuffle_valid_page(unsigned long pfn, int order) +static struct page * __meminit shuffle_valid_page(struct zone *zone, + unsigned long pfn, int order) { - struct page *page; + struct page *page = pfn_to_online_page(pfn); /* * Given we're dealing with randomly selected pfns in a zone we * need to ask questions like... */ - /* ...is the pfn even in the memmap? */ - if (!pfn_valid_within(pfn)) + /* ... is the page managed by the buddy? */ + if (!page) return NULL; - /* ...is the pfn in a present section or a hole? */ - if (!pfn_in_present_section(pfn)) + /* ... is the page assigned to the same zone? */ + if (page_zone(page) != zone) return NULL; /* ...is the page free and currently on a free_area list? */ - page = pfn_to_page(pfn); if (!PageBuddy(page)) return NULL; @@ -123,7 +123,7 @@ void __meminit __shuffle_zone(struct zone *z) * page_j randomly selected in the span @zone_start_pfn to * @spanned_pages. */ - page_i = shuffle_valid_page(i, order); + page_i = shuffle_valid_page(z, i, order); if (!page_i) continue; @@ -137,7 +137,7 @@ void __meminit __shuffle_zone(struct zone *z) j = z->zone_start_pfn + ALIGN_DOWN(get_random_long() % z->spanned_pages, order_pages); - page_j = shuffle_valid_page(j, order); + page_j = shuffle_valid_page(z, j, order); if (page_j && page_j != page_i) break; } -- 2.26.2

4 years, 5 months

2
4
0 0

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror June 2020