February 2019 - Linux-stable-mirror

by Haven Consultants Co

Dear Candidates, Work from home, we have a part time job offer available for you that will not disturb your present occupation, you are to stand as a payment receiving officer for Haven Consultants Co (Thailand) Payment Managers Wanted for 2019/2020. If you are interested in this business transaction, forward to us the Information below to enable the (HRM) process your application as possible. (I)Your Full Names; (II)Mailing address, (III)Phone Number: (IV)Present job; (And Position Held) (V)Age: (VI)Mode of Identification (If Any) (VII)Bank name (VIII)Company name: Please send the above details to this Email below: Email : shigeoueki103(a)gmail.com Regards Shigeo Ueki Haven Consultants Co., Ltd 42/356 Moo 5 Soi Nernplubwan, Nongprue, Banglamung, Chonburi 20150, Thailand Tel 038 196 410 Fax 038 196 411 Email: shigeoueki103(a)gmail.com

5 years, 10 months

1
0
0 0

[PATCH 1/2] MIPS: eBPF: Always return sign extended 32b values

by Paul Burton

The function prototype used to call JITed eBPF code (ie. the type of the struct bpf_prog bpf_func field) returns an unsigned int. The MIPS n64 ABI that MIPS64 kernels target defines that 32 bit integers should always be sign extended when passed in registers as either arguments or return values. This means that when returning any value which may not already be sign extended (ie. of type REG_64BIT or REG_32BIT_ZERO_EX) we need to perform that sign extension in order to comply with the n64 ABI. Without this we see strange looking test failures from test_bpf.ko, such as: test_bpf: #65 ALU64_MOV_X: dst = 4294967295 jited:1 ret -1 != -1 FAIL (1 times) Although the return value printed matches the expected value, this is only because printf is only examining the least significant 32 bits of the 64 bit register value we returned. The register holding the expected value is sign extended whilst the v0 register was set to a zero extended value by our JITed code, so when compared by a conditional branch instruction the values are not equal. We already handle this when the return value register is of type REG_32BIT_ZERO_EX, so simply extend this to also cover REG_64BIT. Signed-off-by: Paul Burton <paul.burton(a)mips.com> Fixes: b6bd53f9c4e8 ("MIPS: Add missing file for eBPF JIT.") Cc: stable(a)vger.kernel.org # v4.13+ --- arch/mips/net/ebpf_jit.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/arch/mips/net/ebpf_jit.c b/arch/mips/net/ebpf_jit.c index b16710a8a9e7..715415fa2345 100644 --- a/arch/mips/net/ebpf_jit.c +++ b/arch/mips/net/ebpf_jit.c @@ -343,12 +343,15 @@ static int build_int_epilogue(struct jit_ctx *ctx, int dest_reg) const struct bpf_prog *prog = ctx->skf; int stack_adjust = ctx->stack_size; int store_offset = stack_adjust - 8; + enum reg_val_type td; int r0 = MIPS_R_V0; - if (dest_reg == MIPS_R_RA && - get_reg_val_type(ctx, prog->len, BPF_REG_0) == REG_32BIT_ZERO_EX) + if (dest_reg == MIPS_R_RA) { /* Don't let zero extended value escape. */ - emit_instr(ctx, sll, r0, r0, 0); + td = get_reg_val_type(ctx, prog->len, BPF_REG_0); + if (td == REG_64BIT || td == REG_32BIT_ZERO_EX) + emit_instr(ctx, sll, r0, r0, 0); + } if (ctx->flags & EBPF_SAVE_RA) { emit_instr(ctx, ld, MIPS_R_RA, store_offset, MIPS_R_SP); -- 2.20.1

5 years, 10 months

1
1
0 0

[stable 4.19 0/4] Intel Wired Ethernet Fixes for 4.19.y

by Jeff Kirsher

This series contains backports and partial backports for the Intel ice driver. These backports are required to get the ice driver to load on the current hardware. This is due to mismatches between the driver code and the firmware running on the devices. At the time of release of 4.19, hardware was not available for verification. Since the 4.19.y kernel is a LTS kernel, we want to ensure that users with our E800 devices will be able to have the in-kernel driver load and pass traffic, if running a 4.19 or later kernel. Anirudh Venkataramanan (4): ice: Update expected FW version ice: Updates to Tx scheduler code ice: Introduce ice_dev_onetime_setup ice: Set timeout when disabling queues .../net/ethernet/intel/ice/ice_adminq_cmd.h | 5 +- drivers/net/ethernet/intel/ice/ice_common.c | 24 +++ drivers/net/ethernet/intel/ice/ice_common.h | 1 + drivers/net/ethernet/intel/ice/ice_controlq.h | 5 +- drivers/net/ethernet/intel/ice/ice_main.c | 2 + drivers/net/ethernet/intel/ice/ice_sched.c | 161 ++++++------------ drivers/net/ethernet/intel/ice/ice_type.h | 2 + 7 files changed, 83 insertions(+), 117 deletions(-) -- 2.20.1

5 years, 10 months

2
10
0 0

[PATCH] libertas_tf: don't set URB_ZERO_PACKET on IN USB transfer

by Lubomir Rintel

It doesn't make sense and the USB core warns on each submit of such URB, easily flooding the message buffer with tracebacks. Analogous issue was fixed in regular libertas driver in commit 6528d8804780 ("libertas: don't set URB_ZERO_PACKET on IN USB transfer"). Cc: stable(a)vger.kernel.org Signed-off-by: Lubomir Rintel <lkundrak(a)v3.sk> --- drivers/net/wireless/marvell/libertas_tf/if_usb.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/wireless/marvell/libertas_tf/if_usb.c b/drivers/net/wireless/marvell/libertas_tf/if_usb.c index 789337ea676a..6ede6168bd85 100644 --- a/drivers/net/wireless/marvell/libertas_tf/if_usb.c +++ b/drivers/net/wireless/marvell/libertas_tf/if_usb.c @@ -433,8 +433,6 @@ static int __if_usb_submit_rx_urb(struct if_usb_card *cardp, skb_tail_pointer(skb), MRVDRV_ETH_RX_PACKET_BUFFER_SIZE, callbackfn, cardp); - cardp->rx_urb->transfer_flags |= URB_ZERO_PACKET; - lbtf_deb_usb2(&cardp->udev->dev, "Pointer for rx_urb %p\n", cardp->rx_urb); ret = usb_submit_urb(cardp->rx_urb, GFP_ATOMIC); -- 2.20.1

5 years, 10 months

3
2
0 0

patch "staging: android: ion: fix sys heap pool's gfp_flags" added to staging-next

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled staging: android: ion: fix sys heap pool's gfp_flags to my staging git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git in the staging-next branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will also be merged in the next major kernel release during the merge window. If you have any questions about this process, please let me know. >From 9bcf065e28122588a6cbee08cf847826dacbb438 Mon Sep 17 00:00:00 2001 From: Qing Xia <saberlily.xia(a)hisilicon.com> Date: Fri, 1 Feb 2019 14:59:46 +0800 Subject: staging: android: ion: fix sys heap pool's gfp_flags In the first loop, gfp_flags will be modified to high_order_gfp_flags, and there will be no chance to change back to low_order_gfp_flags. Fixes: e7f63771b60e ("ION: Sys_heap: Add cached pool to spead up cached buffer alloc") Signed-off-by: Qing Xia <saberlily.xia(a)hisilicon.com> Cc: stable <stable(a)vger.kernel.org> Signed-off-by: Jing Xia <jing.xia(a)unisoc.com> Reviewed-by: Yuming Han <yuming.han(a)unisoc.com> Reviewed-by: Zhaoyang Huang <zhaoyang.huang(a)unisoc.com> Reviewed-by: Orson Zhai <orson.zhai(a)unisoc.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/staging/android/ion/ion_system_heap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/android/ion/ion_system_heap.c b/drivers/staging/android/ion/ion_system_heap.c index ec526a464db8..aa8d8425be25 100644 --- a/drivers/staging/android/ion/ion_system_heap.c +++ b/drivers/staging/android/ion/ion_system_heap.c @@ -224,10 +224,10 @@ static void ion_system_heap_destroy_pools(struct ion_page_pool **pools) static int ion_system_heap_create_pools(struct ion_page_pool **pools) { int i; - gfp_t gfp_flags = low_order_gfp_flags; for (i = 0; i < NUM_ORDERS; i++) { struct ion_page_pool *pool; + gfp_t gfp_flags = low_order_gfp_flags; if (orders[i] > 4) gfp_flags = high_order_gfp_flags; -- 2.20.1

5 years, 10 months

1
0
0 0

[PATCH] mpt3sas: Use driver scsi lookup to track outstanding IOs

by Sreekanth Reddy

Use driver's scsiio lookup table to track outstanding IOs at driver level, Have done following changes in this patch, * Allocated & initialized scsi_lookup table of type (struct scsiio_tracker) and of depth host's can_queue at driver load time and it will be deallocated at driver unload time. * Once scmd is receieved, driver will take scsiio_tracker at entry corresponding to scmd's tag value from scsi_lookup table. Then this scsiio_tracker entry is initialized with proper smid, scmd, cb_idx etc. And this scsiio_tracker entry contents are cleared before driver calling scsi_done callback function. * scmd's host_scribble variable is used to save the corresponding scsiio_tracker address, later at any time driver can easily retrieve the scmd's corresponding scsiio_tacker using this host_scribble variable. * Whenever driver wants to get the outstanding IOs at the driver level then driver can go through this scsi_lookup table and if it observe any entry with non-null scmd then it means that scmd is outstanding at the driver level. Cc: <stable(a)vger.kernel.org> Signed-off-by: Sreekanth Reddy <sreekanth.reddy(a)broadcom.com> --- drivers/scsi/mpt3sas/mpt3sas_base.c | 45 +++++++++++++++++++++++++++++--- drivers/scsi/mpt3sas/mpt3sas_base.h | 10 +++++++ drivers/scsi/mpt3sas/mpt3sas_ctl.c | 2 +- drivers/scsi/mpt3sas/mpt3sas_scsih.c | 16 +++++------- drivers/scsi/mpt3sas/mpt3sas_warpdrive.c | 2 +- 5 files changed, 60 insertions(+), 15 deletions(-) diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.c b/drivers/scsi/mpt3sas/mpt3sas_base.c index 0a6cb8f..8e3d0d5 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_base.c +++ b/drivers/scsi/mpt3sas/mpt3sas_base.c @@ -1301,7 +1301,7 @@ static int mpt3sas_remove_dead_ioc_func(void *arg) cmd = mpt3sas_scsih_scsi_lookup_get(ioc, smid); if (cmd) - return scsi_cmd_priv(cmd); + return mpt3sas_get_st_from_scmd(cmd); return NULL; } @@ -1709,7 +1709,7 @@ static int mpt3sas_remove_dead_ioc_func(void *arg) struct scsi_cmnd *scmd) { struct chain_tracker *chain_req; - struct scsiio_tracker *st = scsi_cmd_priv(scmd); + struct scsiio_tracker *st = mpt3sas_get_st_from_scmd(scmd); u16 smid = st->smid; u8 chain_offset = atomic_read(&ioc->chain_lookup[smid - 1].chain_offset); @@ -3203,14 +3203,18 @@ static int mpt3sas_remove_dead_ioc_func(void *arg) mpt3sas_base_get_smid_scsiio(struct MPT3SAS_ADAPTER *ioc, u8 cb_idx, struct scsi_cmnd *scmd) { - struct scsiio_tracker *request = scsi_cmd_priv(scmd); + struct scsiio_tracker *request; unsigned int tag = scmd->request->tag; u16 smid; + scmd->host_scribble = (unsigned char *)(&ioc->scsi_lookup[tag]); + request = mpt3sas_get_st_from_scmd(scmd); + smid = tag + 1; request->cb_idx = cb_idx; request->msix_io = _base_get_msix_index(ioc); request->smid = smid; + request->scmd = scmd; INIT_LIST_HEAD(&request->chain_list); return smid; } @@ -3264,6 +3268,7 @@ void mpt3sas_base_clear_st(struct MPT3SAS_ADAPTER *ioc, return; st->cb_idx = 0xFF; st->direct_io = 0; + st->scmd = NULL; atomic_set(&ioc->chain_lookup[st->smid - 1].chain_offset, 0); st->smid = 0; } @@ -4252,6 +4257,11 @@ void mpt3sas_base_clear_st(struct MPT3SAS_ADAPTER *ioc, ioc->config_page, ioc->config_page_dma); } + if (ioc->scsi_lookup) { + free_pages((ulong)ioc->scsi_lookup, ioc->scsi_lookup_pages); + ioc->scsi_lookup = NULL; + } + kfree(ioc->hpr_lookup); kfree(ioc->internal_lookup); if (ioc->chain_lookup) { @@ -4377,6 +4387,7 @@ void mpt3sas_base_clear_st(struct MPT3SAS_ADAPTER *ioc, max_request_credit = min_t(u16, facts->RequestCredit, MAX_HBA_QUEUE_DEPTH); +retry: /* Firmware maintains additional facts->HighPriorityCredit number of * credits for HiPriprity Request messages, so hba queue depth will be * sum of max_request_credit and high priority queue depth. @@ -4581,9 +4592,29 @@ void mpt3sas_base_clear_st(struct MPT3SAS_ADAPTER *ioc, (unsigned long long)ioc->request_dma)); total_sz += sz; + sz = ioc->shost->can_queue * sizeof(struct scsiio_tracker); + ioc->scsi_lookup_pages = get_order(sz); + ioc->scsi_lookup = (struct scsiio_tracker *)__get_free_pages( + GFP_KERNEL, ioc->scsi_lookup_pages); + if (!ioc->scsi_lookup) { + /* Retry allocating memory by reducing the queue depth */ + if ((max_request_credit - 64) > + (ioc->internal_depth + INTERNAL_SCSIIO_CMDS_COUNT)) { + max_request_credit -= 64; + _base_release_memory_pools(ioc); + goto retry; + } else { + ioc_err(ioc, + "scsi_lookup: get_free_pages failed, sz(%d)\n", + (int)sz); + goto out; + } + } + dinitprintk(ioc, ioc_info(ioc, "scsiio(0x%p): depth(%d)\n", ioc->request, ioc->scsiio_depth)); + total_sz += sz; ioc->chain_depth = min_t(u32, ioc->chain_depth, MAX_CHAIN_DEPTH); sz = ioc->scsiio_depth * sizeof(struct chain_lookup); @@ -6239,6 +6270,14 @@ void mpt3sas_base_clear_st(struct MPT3SAS_ADAPTER *ioc, spin_lock_irqsave(&ioc->scsi_lookup_lock, flags); + smid = 1; + for (i = 0; i < ioc->shost->can_queue; i++, smid++) { + ioc->scsi_lookup[i].cb_idx = 0xFF; + ioc->scsi_lookup[i].smid = smid; + ioc->scsi_lookup[i].scmd = NULL; + ioc->scsi_lookup[i].direct_io = 0; + } + /* hi-priority queue */ INIT_LIST_HEAD(&ioc->hpr_free_list); smid = ioc->hi_priority_smid; diff --git a/drivers/scsi/mpt3sas/mpt3sas_base.h b/drivers/scsi/mpt3sas/mpt3sas_base.h index 19158cb..f8c82f6 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_base.h +++ b/drivers/scsi/mpt3sas/mpt3sas_base.h @@ -816,6 +816,7 @@ struct chain_lookup { /** * struct scsiio_tracker - scsi mf request tracker * @smid: system message id + * @scmd: scsi request pointer * @cb_idx: callback index * @direct_io: To indicate whether I/O is direct (WARPDRIVE) * @chain_list: list of associated firmware chain tracker @@ -823,6 +824,7 @@ struct chain_lookup { */ struct scsiio_tracker { u16 smid; + struct scsi_cmnd *scmd; u8 cb_idx; u8 direct_io; struct pcie_sg_list pcie_sg_list; @@ -830,6 +832,12 @@ struct scsiio_tracker { u16 msix_io; }; +static inline struct scsiio_tracker * +mpt3sas_get_st_from_scmd(struct scsi_cmnd *scmd) +{ + return (struct scsiio_tracker *)scmd->host_scribble; +} + /** * struct request_tracker - firmware request tracker * @smid: system message id @@ -1296,6 +1304,8 @@ struct MPT3SAS_ADAPTER { u8 *request; dma_addr_t request_dma; u32 request_dma_sz; + struct scsiio_tracker *scsi_lookup; + ulong scsi_lookup_pages; struct pcie_sg_list *pcie_sg_lookup; spinlock_t scsi_lookup_lock; int pending_io_count; diff --git a/drivers/scsi/mpt3sas/mpt3sas_ctl.c b/drivers/scsi/mpt3sas/mpt3sas_ctl.c index b2bb47c..ad43e60 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_ctl.c +++ b/drivers/scsi/mpt3sas/mpt3sas_ctl.c @@ -595,7 +595,7 @@ void mpt3sas_ctl_reset_done_handler(struct MPT3SAS_ADAPTER *ioc) continue; if (priv_data->sas_target->handle != handle) continue; - st = scsi_cmd_priv(scmd); + st = mpt3sas_get_st_from_scmd(scmd); tm_request->TaskMID = cpu_to_le16(st->smid); found = 1; } diff --git a/drivers/scsi/mpt3sas/mpt3sas_scsih.c b/drivers/scsi/mpt3sas/mpt3sas_scsih.c index 8bb5b8f..86d0e3c 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_scsih.c +++ b/drivers/scsi/mpt3sas/mpt3sas_scsih.c @@ -1465,11 +1465,9 @@ struct scsi_cmnd * if (smid > 0 && smid <= ioc->scsiio_depth - INTERNAL_SCSIIO_CMDS_COUNT) { - u32 unique_tag = smid - 1; - - scmd = scsi_host_find_tag(ioc->shost, unique_tag); + scmd = ioc->scsi_lookup[smid - 1].scmd; if (scmd) { - st = scsi_cmd_priv(scmd); + st = mpt3sas_get_st_from_scmd(scmd); if (st->cb_idx == 0xFF || st->smid == 0) scmd = NULL; } @@ -2819,7 +2817,7 @@ int mpt3sas_scsih_issue_locked_tm(struct MPT3SAS_ADAPTER *ioc, u16 handle, { struct MPT3SAS_ADAPTER *ioc = shost_priv(scmd->device->host); struct MPT3SAS_DEVICE *sas_device_priv_data; - struct scsiio_tracker *st = scsi_cmd_priv(scmd); + struct scsiio_tracker *st = mpt3sas_get_st_from_scmd(scmd); u16 handle; int r; @@ -4466,7 +4464,7 @@ static int _scsih_set_satl_pending(struct scsi_cmnd *scmd, bool pending) continue; count++; _scsih_set_satl_pending(scmd, false); - st = scsi_cmd_priv(scmd); + st = mpt3sas_get_st_from_scmd(scmd); mpt3sas_base_clear_st(ioc, st); scsi_dma_unmap(scmd); if (ioc->pci_error_recovery || ioc->remove_host) @@ -5193,7 +5191,7 @@ static int _scsih_set_satl_pending(struct scsi_cmnd *scmd, bool pending) * WARPDRIVE: If direct_io is set then it is directIO, * the failed direct I/O should be redirected to volume */ - st = scsi_cmd_priv(scmd); + st = mpt3sas_get_st_from_scmd(scmd); if (st->direct_io && ((ioc_status & MPI2_IOCSTATUS_MASK) != MPI2_IOCSTATUS_SCSI_TASK_TERMINATED)) { @@ -7335,7 +7333,7 @@ static int _scsih_set_satl_pending(struct scsi_cmnd *scmd, bool pending) scmd = mpt3sas_scsih_scsi_lookup_get(ioc, smid); if (!scmd) continue; - st = scsi_cmd_priv(scmd); + st = mpt3sas_get_st_from_scmd(scmd); sdev = scmd->device; sas_device_priv_data = sdev->hostdata; if (!sas_device_priv_data || !sas_device_priv_data->sas_target) @@ -10176,7 +10174,6 @@ static void pcie_device_make_active(struct MPT3SAS_ADAPTER *ioc, .shost_attrs = mpt3sas_host_attrs, .sdev_attrs = mpt3sas_dev_attrs, .track_queue_depth = 1, - .cmd_size = sizeof(struct scsiio_tracker), }; /* raid transport support for SAS 2.0 HBA devices */ @@ -10214,7 +10211,6 @@ static void pcie_device_make_active(struct MPT3SAS_ADAPTER *ioc, .shost_attrs = mpt3sas_host_attrs, .sdev_attrs = mpt3sas_dev_attrs, .track_queue_depth = 1, - .cmd_size = sizeof(struct scsiio_tracker), }; /* raid transport support for SAS 3.0 HBA devices */ diff --git a/drivers/scsi/mpt3sas/mpt3sas_warpdrive.c b/drivers/scsi/mpt3sas/mpt3sas_warpdrive.c index cc07ba4..2a05bf3 100644 --- a/drivers/scsi/mpt3sas/mpt3sas_warpdrive.c +++ b/drivers/scsi/mpt3sas/mpt3sas_warpdrive.c @@ -259,7 +259,7 @@ sector_t v_lba, p_lba, stripe_off, column, io_size; u32 stripe_sz, stripe_exp; u8 num_pds, cmd = scmd->cmnd[0]; - struct scsiio_tracker *st = scsi_cmd_priv(scmd); + struct scsiio_tracker *st = mpt3sas_get_st_from_scmd(scmd); if (cmd != READ_10 && cmd != WRITE_10 && cmd != READ_16 && cmd != WRITE_16) -- 1.8.3.1

5 years, 10 months

1
0
0 0

[PATCH] PCI: qcom: Don't deassert reset GPIO during probe

by Bjorn Andersson

Acquiring the reset GPIO low means that reset is being deasserted, this is followed almost immediately with qcom_pcie_host_init() asserting it, initializing it and then finally deasserting it again, for the link to come up. Some PCIe devices requires a minimum time between the initial deassert and subsequent reset cycles. In a platform that boots with the reset GPIO asserted this requirement is being violated by this deassert/assert pulse. Acquiring the reset GPIO high will prevent this by matching the state to the subsequent asserted state. Cc: stable(a)vger.kernel.org Signed-off-by: Bjorn Andersson <bjorn.andersson(a)linaro.org> --- drivers/pci/controller/dwc/pcie-qcom.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/controller/dwc/pcie-qcom.c b/drivers/pci/controller/dwc/pcie-qcom.c index d185ea5fe996..a7f703556790 100644 --- a/drivers/pci/controller/dwc/pcie-qcom.c +++ b/drivers/pci/controller/dwc/pcie-qcom.c @@ -1228,7 +1228,7 @@ static int qcom_pcie_probe(struct platform_device *pdev) pcie->ops = of_device_get_match_data(dev); - pcie->reset = devm_gpiod_get_optional(dev, "perst", GPIOD_OUT_LOW); + pcie->reset = devm_gpiod_get_optional(dev, "perst", GPIOD_OUT_HIGH); if (IS_ERR(pcie->reset)) { ret = PTR_ERR(pcie->reset); goto err_pm_runtime_put; -- 2.18.0

5 years, 10 months

3
4
0 0

FAILED: patch "[PATCH] futex: Cure exit race" failed to apply to 4.14-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.14-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From da791a667536bf8322042e38ca85d55a78d3c273 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner <tglx(a)linutronix.de> Date: Mon, 10 Dec 2018 14:35:14 +0100 Subject: [PATCH] futex: Cure exit race Stefan reported, that the glibc tst-robustpi4 test case fails occasionally. That case creates the following race between sys_exit() and sys_futex_lock_pi(): CPU0 CPU1 sys_exit() sys_futex() do_exit() futex_lock_pi() exit_signals(tsk) No waiters: tsk->flags |= PF_EXITING; *uaddr == 0x00000PID mm_release(tsk) Set waiter bit exit_robust_list(tsk) { *uaddr = 0x80000PID; Set owner died attach_to_pi_owner() { *uaddr = 0xC0000000; tsk = get_task(PID); } if (!tsk->flags & PF_EXITING) { ... attach(); tsk->flags |= PF_EXITPIDONE; } else { if (!(tsk->flags & PF_EXITPIDONE)) return -EAGAIN; return -ESRCH; <--- FAIL } ESRCH is returned all the way to user space, which triggers the glibc test case assert. Returning ESRCH unconditionally is wrong here because the user space value has been changed by the exiting task to 0xC0000000, i.e. the FUTEX_OWNER_DIED bit is set and the futex PID value has been cleared. This is a valid state and the kernel has to handle it, i.e. taking the futex. Cure it by rereading the user space value when PF_EXITING and PF_EXITPIDONE is set in the task which 'owns' the futex. If the value has changed, let the kernel retry the operation, which includes all regular sanity checks and correctly handles the FUTEX_OWNER_DIED case. If it hasn't changed, then return ESRCH as there is no way to distinguish this case from malfunctioning user space. This happens when the exiting task did not have a robust list, the robust list was corrupted or the user space value in the futex was simply bogus. Reported-by: Stefan Liebler <stli(a)linux.ibm.com> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Acked-by: Peter Zijlstra <peterz(a)infradead.org> Cc: Heiko Carstens <heiko.carstens(a)de.ibm.com> Cc: Darren Hart <dvhart(a)infradead.org> Cc: Ingo Molnar <mingo(a)kernel.org> Cc: Sasha Levin <sashal(a)kernel.org> Cc: stable(a)vger.kernel.org Link: https://bugzilla.kernel.org/show_bug.cgi?id=200467 Link: https://lkml.kernel.org/r/20181210152311.986181245@linutronix.de diff --git a/kernel/futex.c b/kernel/futex.c index f423f9b6577e..5cc8083a4c89 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -1148,11 +1148,65 @@ static int attach_to_pi_state(u32 __user *uaddr, u32 uval, return ret; } +static int handle_exit_race(u32 __user *uaddr, u32 uval, + struct task_struct *tsk) +{ + u32 uval2; + + /* + * If PF_EXITPIDONE is not yet set, then try again. + */ + if (tsk && !(tsk->flags & PF_EXITPIDONE)) + return -EAGAIN; + + /* + * Reread the user space value to handle the following situation: + * + * CPU0 CPU1 + * + * sys_exit() sys_futex() + * do_exit() futex_lock_pi() + * futex_lock_pi_atomic() + * exit_signals(tsk) No waiters: + * tsk->flags |= PF_EXITING; *uaddr == 0x00000PID + * mm_release(tsk) Set waiter bit + * exit_robust_list(tsk) { *uaddr = 0x80000PID; + * Set owner died attach_to_pi_owner() { + * *uaddr = 0xC0000000; tsk = get_task(PID); + * } if (!tsk->flags & PF_EXITING) { + * ... attach(); + * tsk->flags |= PF_EXITPIDONE; } else { + * if (!(tsk->flags & PF_EXITPIDONE)) + * return -EAGAIN; + * return -ESRCH; <--- FAIL + * } + * + * Returning ESRCH unconditionally is wrong here because the + * user space value has been changed by the exiting task. + * + * The same logic applies to the case where the exiting task is + * already gone. + */ + if (get_futex_value_locked(&uval2, uaddr)) + return -EFAULT; + + /* If the user space value has changed, try again. */ + if (uval2 != uval) + return -EAGAIN; + + /* + * The exiting task did not have a robust list, the robust list was + * corrupted or the user space value in *uaddr is simply bogus. + * Give up and tell user space. + */ + return -ESRCH; +} + /* * Lookup the task for the TID provided from user space and attach to * it after doing proper sanity checks. */ -static int attach_to_pi_owner(u32 uval, union futex_key *key, +static int attach_to_pi_owner(u32 __user *uaddr, u32 uval, union futex_key *key, struct futex_pi_state **ps) { pid_t pid = uval & FUTEX_TID_MASK; @@ -1162,12 +1216,15 @@ static int attach_to_pi_owner(u32 uval, union futex_key *key, /* * We are the first waiter - try to look up the real owner and attach * the new pi_state to it, but bail out when TID = 0 [1] + * + * The !pid check is paranoid. None of the call sites should end up + * with pid == 0, but better safe than sorry. Let the caller retry */ if (!pid) - return -ESRCH; + return -EAGAIN; p = find_get_task_by_vpid(pid); if (!p) - return -ESRCH; + return handle_exit_race(uaddr, uval, NULL); if (unlikely(p->flags & PF_KTHREAD)) { put_task_struct(p); @@ -1187,7 +1244,7 @@ static int attach_to_pi_owner(u32 uval, union futex_key *key, * set, we know that the task has finished the * cleanup: */ - int ret = (p->flags & PF_EXITPIDONE) ? -ESRCH : -EAGAIN; + int ret = handle_exit_race(uaddr, uval, p); raw_spin_unlock_irq(&p->pi_lock); put_task_struct(p); @@ -1244,7 +1301,7 @@ static int lookup_pi_state(u32 __user *uaddr, u32 uval, * We are the first waiter - try to look up the owner based on * @uval and attach to it. */ - return attach_to_pi_owner(uval, key, ps); + return attach_to_pi_owner(uaddr, uval, key, ps); } static int lock_pi_update_atomic(u32 __user *uaddr, u32 uval, u32 newval) @@ -1352,7 +1409,7 @@ static int futex_lock_pi_atomic(u32 __user *uaddr, struct futex_hash_bucket *hb, * attach to the owner. If that fails, no harm done, we only * set the FUTEX_WAITERS bit in the user space variable. */ - return attach_to_pi_owner(uval, key, ps); + return attach_to_pi_owner(uaddr, newval, key, ps); } /**

5 years, 10 months

5
7
0 0

patch "staging: android: ion: fix sys heap pool's gfp_flags" added to staging-testing

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled staging: android: ion: fix sys heap pool's gfp_flags to my staging git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging.git in the staging-testing branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will be merged to the staging-next branch sometime soon, after it passes testing, and the merge window is open. If you have any questions about this process, please let me know. >From 9bcf065e28122588a6cbee08cf847826dacbb438 Mon Sep 17 00:00:00 2001 From: Qing Xia <saberlily.xia(a)hisilicon.com> Date: Fri, 1 Feb 2019 14:59:46 +0800 Subject: staging: android: ion: fix sys heap pool's gfp_flags In the first loop, gfp_flags will be modified to high_order_gfp_flags, and there will be no chance to change back to low_order_gfp_flags. Fixes: e7f63771b60e ("ION: Sys_heap: Add cached pool to spead up cached buffer alloc") Signed-off-by: Qing Xia <saberlily.xia(a)hisilicon.com> Cc: stable <stable(a)vger.kernel.org> Signed-off-by: Jing Xia <jing.xia(a)unisoc.com> Reviewed-by: Yuming Han <yuming.han(a)unisoc.com> Reviewed-by: Zhaoyang Huang <zhaoyang.huang(a)unisoc.com> Reviewed-by: Orson Zhai <orson.zhai(a)unisoc.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/staging/android/ion/ion_system_heap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/android/ion/ion_system_heap.c b/drivers/staging/android/ion/ion_system_heap.c index ec526a464db8..aa8d8425be25 100644 --- a/drivers/staging/android/ion/ion_system_heap.c +++ b/drivers/staging/android/ion/ion_system_heap.c @@ -224,10 +224,10 @@ static void ion_system_heap_destroy_pools(struct ion_page_pool **pools) static int ion_system_heap_create_pools(struct ion_page_pool **pools) { int i; - gfp_t gfp_flags = low_order_gfp_flags; for (i = 0; i < NUM_ORDERS; i++) { struct ion_page_pool *pool; + gfp_t gfp_flags = low_order_gfp_flags; if (orders[i] > 4) gfp_flags = high_order_gfp_flags; -- 2.20.1

5 years, 10 months

1
0
0 0

[PATCH AUTOSEL 3.18 01/16] drm/msm: Unblock writer if reader closes file

by Sasha Levin

From: "Kristian H. Kristensen" <hoegsberg(a)gmail.com> [ Upstream commit 99c66bc051e7407fe0bf0607b142ec0be1a1d1dd ] Prevents deadlock when fifo is full and reader closes file. Signed-off-by: Kristian H. Kristensen <hoegsberg(a)chromium.org> Signed-off-by: Rob Clark <robdclark(a)gmail.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- drivers/gpu/drm/msm/msm_rd.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index 9a78c48817c6..909a52b21ebe 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -103,7 +103,9 @@ static void rd_write(struct msm_rd_state *rd, const void *buf, int sz) char *fptr = &fifo->buf[fifo->head]; int n; - wait_event(rd->fifo_event, circ_space(&rd->fifo) > 0); + wait_event(rd->fifo_event, circ_space(&rd->fifo) > 0 || !rd->open); + if (!rd->open) + return; n = min(sz, circ_space_to_end(&rd->fifo)); memcpy(fptr, ptr, n); @@ -192,7 +194,10 @@ static int rd_open(struct inode *inode, struct file *file) static int rd_release(struct inode *inode, struct file *file) { struct msm_rd_state *rd = inode->i_private; + rd->open = false; + wake_up_all(&rd->fifo_event); + return 0; } -- 2.19.1

5 years, 10 months

3
17
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror February 2019