December 2022 - Kernel - mailweb.openeuler.org

[PATCH openEuler-1.0-LTS] mm/sharepool: Charge Buddy hugepage to memcg
by Yongqiang Liu 30 Dec '22

30 Dec '22

From: Zhou Guanghui <zhouguanghui1(a)huawei.com> ascend inclusion category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I67NC1 CVE: NA ------------------------------------------ Charge Buddy hugepage to memcg when kmemcg is disabled. If kmemcg is enabled, we can alse use kmemcg to charge buddy hugepages. Signed-off-by: Zhou Guanghui <zhouguanghui1(a)huawei.com> Reviewed-by: Kefeng Wang <wangkefeng.wang(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- include/linux/share_pool.h | 30 ++++++++++++++++++++++++++++++ mm/hugetlb.c | 3 +++ mm/share_pool.c | 28 +++++++++++++++++++++++----- 3 files changed, 56 insertions(+), 5 deletions(-) diff --git a/include/linux/share_pool.h b/include/linux/share_pool.h index c58f0dce0b20..2ebfb3d5be13 100644 --- a/include/linux/share_pool.h +++ b/include/linux/share_pool.h @@ -8,6 +8,8 @@ #include <linux/printk.h> #include <linux/hashtable.h> #include <linux/numa.h> +#include <linux/hugetlb.h> +#include <linux/memcontrol.h> #define SP_HUGEPAGE (1 << 0) #define SP_HUGEPAGE_ONLY (1 << 1) @@ -411,6 +413,26 @@ extern bool sp_check_mmap_addr(unsigned long addr, unsigned long flags); extern int sp_id_of_current(void); extern int mg_sp_id_of_current(void); + +static inline void sp_kmemcg_uncharge_hpage(struct page *page) +{ + if (!sp_is_enabled()) + return; + + if (memcg_kmem_enabled() && PageKmemcg(page)) { + int order = huge_page_order(page_hstate(page)); + + __memcg_kmem_uncharge(page, order); + } +} + +static inline void sp_memcg_uncharge_hpage(struct page *page) +{ + if (!sp_is_enabled()) + return; + + mem_cgroup_uncharge(page); +} #else static inline int mg_sp_group_add_task(int pid, unsigned long prot, int spg_id) @@ -684,6 +706,14 @@ static inline int mg_sp_id_of_current(void) return -EPERM; } +static inline void sp_kmemcg_uncharge_hpage(struct page *page) +{ +} + +static inline void sp_memcg_uncharge_hpage(struct page *page) +{ +} + #endif #endif /* LINUX_SHARE_POOL_H */ diff --git a/mm/hugetlb.c b/mm/hugetlb.c index fb1f90eed7fc..495d8b5b38fc 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -38,6 +38,7 @@ #include <linux/node.h> #include <linux/userfaultfd_k.h> #include <linux/page_owner.h> +#include <linux/share_pool.h> #include "internal.h" int hugetlb_max_hstate __read_mostly; @@ -1311,6 +1312,7 @@ void free_huge_page(struct page *page) (struct hugepage_subpool *)page_private(page); bool restore_reserve; + sp_kmemcg_uncharge_hpage(page); set_page_private(page, 0); page->mapping = NULL; VM_BUG_ON_PAGE(page_count(page), page); @@ -1345,6 +1347,7 @@ void free_huge_page(struct page *page) h->resv_huge_pages++; if (PageHugeTemporary(page)) { + sp_memcg_uncharge_hpage(page); list_del(&page->lru); ClearPageHugeTemporary(page); update_and_free_page(h, page); diff --git a/mm/share_pool.c b/mm/share_pool.c index d2591c73a503..5d8344fe805a 100644 --- a/mm/share_pool.c +++ b/mm/share_pool.c @@ -4555,6 +4555,8 @@ vm_fault_t sharepool_no_page(struct mm_struct *mm, int err; int node_id; struct sp_area *spa; + bool charge_hpage = false; + struct mem_cgroup *memcg; spa = vma->vm_private_data; if (!spa) { @@ -4572,10 +4574,11 @@ vm_fault_t sharepool_no_page(struct mm_struct *mm, page = alloc_huge_page(vma, haddr, 0); if (IS_ERR(page)) { - page = alloc_huge_page_node(hstate_file(vma->vm_file), - node_id); + page = hugetlb_alloc_hugepage(node_id, HUGETLB_ALLOC_BUDDY); if (!page) page = ERR_PTR(-ENOMEM); + else if (!PageKmemcg(page)) + charge_hpage = true; } if (IS_ERR(page)) { ptl = huge_pte_lock(h, mm, ptep); @@ -4588,12 +4591,24 @@ vm_fault_t sharepool_no_page(struct mm_struct *mm, ret = vmf_error(PTR_ERR(page)); goto out; } + + if (charge_hpage && + mem_cgroup_try_charge_delay(page, vma->vm_mm, GFP_KERNEL, &memcg, true)) { + put_page(page); + ret = vmf_error(-ENOMEM); + goto out; + } + __SetPageUptodate(page); new_page = true; /* sharepool pages are all shared */ err = huge_add_to_page_cache(page, mapping, idx); if (err) { + if (charge_hpage) { + mem_cgroup_cancel_charge(page, memcg, true); + charge_hpage = false; + } put_page(page); if (err == -EEXIST) goto retry; @@ -4601,7 +4616,6 @@ vm_fault_t sharepool_no_page(struct mm_struct *mm, } } - ptl = huge_pte_lock(h, mm, ptep); size = i_size_read(mapping->host) >> huge_page_shift(h); if (idx >= size) @@ -4618,11 +4632,13 @@ vm_fault_t sharepool_no_page(struct mm_struct *mm, hugetlb_count_add(pages_per_huge_page(h), mm); + if (charge_hpage) + mem_cgroup_commit_charge(page, memcg, false, true); + spin_unlock(ptl); - if (new_page) { + if (new_page) SetPagePrivate(&page[1]); - } unlock_page(page); out: @@ -4631,6 +4647,8 @@ vm_fault_t sharepool_no_page(struct mm_struct *mm, backout: spin_unlock(ptl); unlock_page(page); + if (charge_hpage) + mem_cgroup_cancel_charge(page, memcg, true); put_page(page); goto out; } -- 2.25.1

1 0

openEuler Kernel SIG双周例会
by openEuler conference 30 Dec '22

30 Dec '22

您好！ Kernel SIG 邀请您参加 2022-12-30 14:00 召开的Zoom会议(自动录制) 会议主题：openEuler Kernel SIG双周例会会议内容： 1. 进展update 2. 议题征集中欢迎大家积极申报议题（新增议题可以回复邮件反馈，或者录入会议看板）会议链接：https://us06web.zoom.us/j/84061666239?pwd=ZGNrZUo0RW5DWWhOQTRaVW53Y0gyUT09 会议纪要：https://etherpad.openeuler.org/p/Kernel-meetings 温馨提醒：建议接入会议后修改参会人的姓名，也可以使用您在gitee.com的ID 更多资讯尽在：https://openeuler.org/zh/ Hello! openEuler Kernel SIG invites you to attend the Zoom conference(auto recording) will be held at 2022-12-30 14:00, The subject of the conference is openEuler Kernel SIG双周例会, Summary: 1. 进展update 2. 议题征集中欢迎大家积极申报议题（新增议题可以回复邮件反馈，或者录入会议看板） You can join the meeting at https://us06web.zoom.us/j/84061666239?pwd=ZGNrZUo0RW5DWWhOQTRaVW53Y0gyUT09. Add topics at https://etherpad.openeuler.org/p/Kernel-meetings. Note: You are advised to change the participant name after joining the conference or use your ID at gitee.com. More information: https://openeuler.org/en/

2 1

[PATCH v2 OLK-5.10] Support 3SNIC 3S5XX serial RAID/HBA controllers
by Steven Song 30 Dec '22

30 Dec '22

3SNIC inclusion category: features bugzilla: https://gitee.com/openeuler/kernel/issues/I64M90 CVE: NA v1->v2: 1. Update openEuler/MAINTAINERS This commit is to suppport 3SNIC 3S5XX RAID/HBA controllers. RAID controllers support RAID 0/1/5/6/10/50/60 modes; HBA controlllers support RAID 0/1/10 modes. RAID/HBA support SAS/SATA HDD/SSD. Signed-off-by: liangry <liangry1(a)3snic.com> Reviewed-by: Jiang Yu<yujiang(a)3snic.com> Reviewed-by: Steven Song <steven.song(a)3snic.com> --- Documentation/scsi/sssraid.rst | 83 + MAINTAINERS | 7 + arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + drivers/scsi/Kconfig | 1 + drivers/scsi/Makefile | 1 + drivers/scsi/sssraid/Kconfig | 14 + drivers/scsi/sssraid/Makefile | 7 + drivers/scsi/sssraid/sssraid.h | 971 ++++++++++ drivers/scsi/sssraid/sssraid_debug.h | 34 + drivers/scsi/sssraid/sssraid_fw.c | 1724 +++++++++++++++++ drivers/scsi/sssraid/sssraid_os.c | 2337 ++++++++++++++++++++++++ openEuler/MAINTAINERS | 6 + 13 files changed, 5187 insertions(+) create mode 100644 Documentation/scsi/sssraid.rst create mode 100644 drivers/scsi/sssraid/Kconfig create mode 100644 drivers/scsi/sssraid/Makefile create mode 100644 drivers/scsi/sssraid/sssraid.h create mode 100644 drivers/scsi/sssraid/sssraid_debug.h create mode 100644 drivers/scsi/sssraid/sssraid_fw.c create mode 100644 drivers/scsi/sssraid/sssraid_os.c diff --git a/Documentation/scsi/sssraid.rst b/Documentation/scsi/sssraid.rst new file mode 100644 index 000000000000..d56e5f7c478c --- /dev/null +++ b/Documentation/scsi/sssraid.rst @@ -0,0 +1,83 @@ +.. SPDX-License-Identifier: GPL-2.0 + +============================================== +SSSRAID - 3SNIC SCSI RAID Controller driver +============================================== + +This file describes the SSSRAID SCSI driver for 3SNIC +(http://www.3snic.com) RAID controllers. The SSSRAID +driver is the first generation RAID driver for 3SNIC Corp. + +For 3SNIC SSSRAID controller support, enable the SSSRAID driver +when configuring the kernel. + +SSSRAID specific entries in /sys +================================= + +SSSRAID host attributes +------------------------ + - /sys/class/scsi_host/host*/csts_pp + - /sys/class/scsi_host/host*/csts_shst + - /sys/class/scsi_host/host*/csts_cfs + - /sys/class/scsi_host/host*/csts_rdy + - /sys/class/scsi_host/host*/fw_version + + The host csts_pp attribute is a read only attribute. This attribute + indicates whether the controller is processing commands. If this attribute + is set to ‘1’, then the controller is processing commands normally. If + this attribute is cleared to ‘0’, then the controller has temporarily stopped + processing commands in order to handle an event (e.g., firmware activation). + + The host csts_shst attribute is a read only attribute. This attribute + indicates status of shutdown processing.The shutdown status values are defined + as: + ====== ============================== + Value Definition + ====== ============================== + 00b Normal operation + 01b Shutdown processing occurring + 10b Shutdown processing complete + 11b Reserved + ====== ============================== + The host csts_cfs attribute is a read only attribute. This attribute is set to + ’1’ when a fatal controller error occurred that could not be communicated in the + appropriate Completion Queue. This bit is cleared to ‘0’ when a fatal controller + error has not occurred. + + The host csts_rdy attribute is a read only attribute. This attribute is set to + ‘1’ when the controller is ready to process submission queue entries. + + The fw_version attribute is read-only and will return the driver version and the + controller firmware version. + +SSSRAID scsi device attributes +------------------------------ + - /sys/class/scsi_device/X\:X\:X\:X/device/raid_level + - /sys/class/scsi_device/X\:X\:X\:X/device/raid_state + - /sys/class/scsi_device/X\:X\:X\:X/device/raid_resync + + The device raid_level attribute is a read only attribute. This attribute indicates + RAID level of scsi device(will dispaly "NA" if scsi device is not virtual disk type). + + The device raid_state attribute is read-only and indicates RAID status of scsi + device(will dispaly "NA" if scsi device is not virtual disk type). + + The device raid_resync attribute is read-only and indicates RAID rebuild processing + of scsi device(will dispaly "NA" if scsi device is not virtual disk type). + +Supported devices +================= + + =================== ======= ======================================= + PCI ID (pci.ids) OEM Product + =================== ======= ======================================= + 1F3F:2100 3SNIC 3S510(HBA:8Ports,1G DDR) + 1F3F:2100 3SNIC 3S520(HBA:16Ports,1G DDR) + 1F3F:2100 3SNIC 3S530(HBA:32Ports,1G DDR) + 1F3F:2100 3SNIC 3S540(HBA:40Ports,1G DDR) + 1F3F:2200 3SNIC 3S580(RAID:16Ports,2G cache) + 1F3F:2200 3SNIC 3S585(RAID:16Ports,4G cache) + 1F3F:2200 3SNIC 3S590(RAID:32Ports,4G cache) + 1F3F:2200 3SNIC 3S5A0(RAID:40Ports,2G cache) + 1F3F:2200 3SNIC 3S5A5(RAID:40Ports,4G cache) + =================== ======= ======================================= diff --git a/MAINTAINERS b/MAINTAINERS index 9ae209633628..e6470640f764 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16669,6 +16669,13 @@ M: Jan-Benedict Glaw <jbglaw(a)lug-owl.de> S: Maintained F: arch/alpha/kernel/srm_env.c +SSSRAID SCSI/Raid DRIVERS +M: Steven Song <steven.song(a)3snic.com> +L: linux-scsi(a)vger.kernel.org +S: Maintained +F: Documentation/scsi/sssraid.rst +F: drivers/scsi/sssraid/ + ST LSM6DSx IMU IIO DRIVER M: Lorenzo Bianconi <lorenzo.bianconi83(a)gmail.com> L: linux-iio(a)vger.kernel.org diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 30384eacad4f..dc4ba0d2b694 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -2394,6 +2394,7 @@ CONFIG_SCSI_HISI_SAS_DEBUGFS_DEFAULT_ENABLE=y # CONFIG_MEGARAID_NEWGEN is not set # CONFIG_MEGARAID_LEGACY is not set CONFIG_MEGARAID_SAS=m +CONFIG_SCSI_3SNIC_SSSRAID=m CONFIG_SCSI_MPT3SAS=m CONFIG_SCSI_MPT2SAS_MAX_SGE=128 CONFIG_SCSI_MPT3SAS_MAX_SGE=128 diff --git a/arch/x86/configs/openeuler_defconfig b/arch/x86/configs/openeuler_defconfig index 5ada612f1d75..df16f123515b 100644 --- a/arch/x86/configs/openeuler_defconfig +++ b/arch/x86/configs/openeuler_defconfig @@ -2355,6 +2355,7 @@ CONFIG_SCSI_AACRAID=m # CONFIG_MEGARAID_LEGACY is not set CONFIG_MEGARAID_SAS=m CONFIG_SCSI_MPT3SAS=m +CONFIG_SCSI_3SNIC_SSSRAID=m CONFIG_SCSI_MPT2SAS_MAX_SGE=128 CONFIG_SCSI_MPT3SAS_MAX_SGE=128 CONFIG_SCSI_MPT2SAS=m diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index 0fbe4edeccd0..7339d6179e5a 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -481,6 +481,7 @@ config SCSI_ARCMSR source "drivers/scsi/esas2r/Kconfig" source "drivers/scsi/megaraid/Kconfig.megaraid" +source "drivers/scsi/sssraid/Kconfig" source "drivers/scsi/mpt3sas/Kconfig" source "drivers/scsi/smartpqi/Kconfig" source "drivers/scsi/ufs/Kconfig" diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile index 78a3c832394c..01bb908fb832 100644 --- a/drivers/scsi/Makefile +++ b/drivers/scsi/Makefile @@ -92,6 +92,7 @@ obj-$(CONFIG_SCSI_CHELSIO_FCOE) += csiostor/ obj-$(CONFIG_SCSI_DMX3191D) += dmx3191d.o obj-$(CONFIG_SCSI_HPSA) += hpsa.o obj-$(CONFIG_SCSI_SMARTPQI) += smartpqi/ +obj-$(CONFIG_SCSI_3SNIC_SSSRAID) += sssraid/ obj-$(CONFIG_SCSI_SYM53C8XX_2) += sym53c8xx_2/ obj-$(CONFIG_SCSI_ZALON) += zalon7xx.o obj-$(CONFIG_SCSI_DC395x) += dc395x.o diff --git a/drivers/scsi/sssraid/Kconfig b/drivers/scsi/sssraid/Kconfig new file mode 100644 index 000000000000..769f68732bf3 --- /dev/null +++ b/drivers/scsi/sssraid/Kconfig @@ -0,0 +1,14 @@ +# +# Kernel configuration file for the 3SNIC +# + +config SCSI_3SNIC_SSSRAID + tristate "3SNIC sssraid Adapter" + depends on PCI && SCSI + select BLK_DEV_BSGLIB + depends on ARM64 || X86_64 + help + This driver supports 3SNIC 3S5xx serial RAID controller, which has + PCI Express Gen4 interface with host and supports SAS/SATA HDD/SSD. + To compile this driver as a module, choose M here: the module will + be called sssraid. diff --git a/drivers/scsi/sssraid/Makefile b/drivers/scsi/sssraid/Makefile new file mode 100644 index 000000000000..c80605866b42 --- /dev/null +++ b/drivers/scsi/sssraid/Makefile @@ -0,0 +1,7 @@ +# +# Makefile for the 3SNIC sssraid drivers. +# + +obj-$(CONFIG_SCSI_3SNIC_SSSRAID) += sssraid.o + +sssraid-objs := sssraid_os.o sssraid_fw.o diff --git a/drivers/scsi/sssraid/sssraid.h b/drivers/scsi/sssraid/sssraid.h new file mode 100644 index 000000000000..bae668ef9c98 --- /dev/null +++ b/drivers/scsi/sssraid/sssraid.h @@ -0,0 +1,971 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2022 3SNIC Information Technology, Ltd */ + +/* 3SNIC RAID SSSXXX Series Linux Driver */ + +#ifndef __SSSRAID_H_ +#define __SSSRAID_H_ + +#define SSSRAID_DRIVER_VERSION "1.0.0.0" +#define SSSRAID_DRIVER_RELDATE "12-December-2022" + +#define SSSRAID_DRIVER_NAME "sssraid" + +#define SSSRAID_NAME_LENGTH 32 + +/* + * SSSRAID Vendor ID and Device IDs + */ +#define PCI_VENDOR_ID_3SNIC_LOGIC 0x1F3F + +#define SSSRAID_SERVER_DEVICE_HBA_DID 0x2100 +#define SSSRAID_SERVER_DEVICE_RAID_DID 0x2200 + +#define SSSRAID_CAP_MQES(cap) ((cap) & 0xffff) +#define SSSRAID_CAP_STRIDE(cap) (((cap) >> 32) & 0xf) +#define SSSRAID_CAP_MPSMIN(cap) (((cap) >> 48) & 0xf) +#define SSSRAID_CAP_MPSMAX(cap) (((cap) >> 52) & 0xf) +#define SSSRAID_CAP_TIMEOUT(cap) (((cap) >> 24) & 0xff) +#define SSSRAID_CAP_DMAMASK(cap) (((cap) >> 37) & 0xff) + +#define SSSRAID_DEFAULT_MAX_CHANNEL 4 +#define SSSRAID_DEFAULT_MAX_ID 240 +#define SSSRAID_DEFAULT_MAX_LUN_PER_HOST 8 +#define MAX_SECTORS 2048 + +/* + * Time define + */ +#define SSSRAID_WATCHDOG_INTERVAL 1000 /* in milli seconds */ +#define SSSRAID_PORTENABLE_TIMEOUT 300 + +/* + * SSSRAID queue and entry size for Admin and I/O type + */ +#define IO_SQE_SIZE sizeof(struct sssraid_ioq_command) +#define ADMIN_SQE_SIZE sizeof(struct sssraid_admin_command) +#define SQE_SIZE(qid) (((qid) > 0) ? IO_SQE_SIZE : ADMIN_SQE_SIZE) +#define CQ_SIZE(depth) ((depth) * sizeof(struct sssraid_completion)) +#define SQ_SIZE(qid, depth) ((depth) * SQE_SIZE(qid)) + +#define SENSE_SIZE(depth) ((depth) * SCSI_SENSE_BUFFERSIZE) + +#define SSSRAID_AQ_DEPTH 128 +#define SSSRAID_NR_AEN_COMMANDS 16 +#define SSSRAID_AQ_BLK_MQ_DEPTH (SSSRAID_AQ_DEPTH - SSSRAID_NR_AEN_COMMANDS) +#define SSSRAID_AQ_MQ_TAG_DEPTH (SSSRAID_AQ_BLK_MQ_DEPTH - 1) + +#define SSSRAID_ADMIN_QUEUE_NUM 1 +#define SSSRAID_PTCMDS_PERQ 1 +#define SSSRAID_IO_BLK_MQ_DEPTH (hdev->scsi_qd) +#define SSSRAID_NR_IOQ_PTCMDS (SSSRAID_PTCMDS_PERQ * hdev->shost->nr_hw_queues) + +#define FUA_MASK 0x08 +#define SSSRAID_MINORS BIT(MINORBITS) +#define SSSRAID_RW_FUA BIT(14) + +#define COMMAND_IS_WRITE(cmd) ((cmd)->common.opcode & 1) + +#define SSSRAID_IO_IOSQES 7 +#define SSSRAID_IO_IOCQES 4 +#define PRP_ENTRY_SIZE 8 + +#define SMALL_POOL_SIZE 256 +#define MAX_SMALL_POOL_NUM 16 +#define MAX_CMD_PER_DEV 64 +#define MAX_CDB_LEN 16 + +#define SSSRAID_UP_TO_MULTY4(x) (((x) + 4) & (~0x03)) + +#define CQE_STATUS_SUCCESS (0x0) + +#define IO_6_DEFAULT_TX_LEN 256 + +#define SGES_PER_PAGE (PAGE_SIZE / sizeof(struct sssraid_sgl_desc)) + +#define SSSRAID_CAP_TIMEOUT_UNIT_MS (HZ / 2) + +extern u32 admin_tmout; +#define ADMIN_TIMEOUT (admin_tmout * HZ) + +#define SSSRAID_WAIT_ABNL_CMD_TIMEOUT 6 + +#define SSSRAID_DMA_MSK_BIT_MAX 64 + +enum { + SSSRAID_SGL_FMT_DATA_DESC = 0x00, + SSSRAID_SGL_FMT_SEG_DESC = 0x02, + SSSRAID_SGL_FMT_LAST_SEG_DESC = 0x03, + SSSRAID_KEY_SGL_FMT_DATA_DESC = 0x04, + SSSRAID_TRANSPORT_SGL_DATA_DESC = 0x05 +}; + + +enum { + SSSRAID_REQ_CANCELLED = (1 << 0), + SSSRAID_REQ_USERCMD = (1 << 1), +}; + +enum { + SSSRAID_SC_SUCCESS = 0x0, + SSSRAID_SC_INVALID_OPCODE = 0x1, + SSSRAID_SC_INVALID_FIELD = 0x2, + + SSSRAID_SC_ABORT_LIMIT = 0x103, + SSSRAID_SC_ABORT_MISSING = 0x104, + SSSRAID_SC_ASYNC_LIMIT = 0x105, + + SSSRAID_SC_DNR = 0x4000, +}; + +enum { + SSSRAID_REG_CAP = 0x0000, + SSSRAID_REG_CC = 0x0014, + SSSRAID_REG_CSTS = 0x001c, + SSSRAID_REG_AQA = 0x0024, + SSSRAID_REG_ASQ = 0x0028, + SSSRAID_REG_ACQ = 0x0030, + SSSRAID_REG_DBS = 0x1000, +}; + +enum { + SSSRAID_CC_ENABLE = 1 << 0, + SSSRAID_CC_CSS_NVM = 0 << 4, + SSSRAID_CC_MPS_SHIFT = 7, + SSSRAID_CC_AMS_SHIFT = 11, + SSSRAID_CC_SHN_SHIFT = 14, + SSSRAID_CC_IOSQES_SHIFT = 16, + SSSRAID_CC_IOCQES_SHIFT = 20, + SSSRAID_CC_AMS_RR = 0 << SSSRAID_CC_AMS_SHIFT, + SSSRAID_CC_SHN_NONE = 0 << SSSRAID_CC_SHN_SHIFT, + SSSRAID_CC_IOSQES = SSSRAID_IO_IOSQES << SSSRAID_CC_IOSQES_SHIFT, + SSSRAID_CC_IOCQES = SSSRAID_IO_IOCQES << SSSRAID_CC_IOCQES_SHIFT, + SSSRAID_CC_SHN_NORMAL = 1 << SSSRAID_CC_SHN_SHIFT, + SSSRAID_CC_SHN_MASK = 3 << SSSRAID_CC_SHN_SHIFT, + SSSRAID_CSTS_CFS_SHIFT = 1, + SSSRAID_CSTS_SHST_SHIFT = 2, + SSSRAID_CSTS_PP_SHIFT = 5, + SSSRAID_CSTS_RDY = 1 << 0, + SSSRAID_CSTS_SHST_CMPLT = 2 << 2, + SSSRAID_CSTS_SHST_MASK = 3 << 2, + SSSRAID_CSTS_CFS_MASK = 1 << SSSRAID_CSTS_CFS_SHIFT, + SSSRAID_CSTS_PP_MASK = 1 << SSSRAID_CSTS_PP_SHIFT, +}; + +enum { + SSSRAID_ADMIN_DELETE_SQ = 0x00, + SSSRAID_ADMIN_CREATE_SQ = 0x01, + SSSRAID_ADMIN_DELETE_CQ = 0x04, + SSSRAID_ADMIN_CREATE_CQ = 0x05, + SSSRAID_ADMIN_ABORT_CMD = 0x08, + SSSRAID_ADMIN_SET_FEATURES = 0x09, + SSSRAID_ADMIN_ASYNC_EVENT = 0x0c, + SSSRAID_ADMIN_GET_INFO = 0xc6, + SSSRAID_ADMIN_RESET = 0xc8, +}; + +enum { + SSSRAID_GET_INFO_CTRL = 0, + SSSRAID_GET_INFO_DEV_LIST = 1, +}; + +enum sssraid_scsi_rst_type { + SSSRAID_RESET_TARGET = 0, + SSSRAID_RESET_BUS = 1, +}; + +enum { + SSSRAID_AEN_ERROR = 0, + SSSRAID_AEN_NOTICE = 2, + SSSRAID_AEN_VS = 7, +}; + +enum { + SSSRAID_AEN_DEV_CHANGED = 0x00, + SSSRAID_AEN_FW_ACT_START = 0x01, + SSSRAID_AEN_HOST_PROBING = 0x10, +}; + +enum { + SSSRAID_AEN_TIMESYN = 0x00, + SSSRAID_AEN_FW_ACT_FINISH = 0x02, + SSSRAID_AEN_EVENT_MIN = 0x80, + SSSRAID_AEN_EVENT_MAX = 0xff, +}; + +enum { + SSSRAID_CMD_WRITE = 0x01, + SSSRAID_CMD_READ = 0x02, + + SSSRAID_CMD_NONIO_NONE = 0x80, + SSSRAID_CMD_NONIO_TODEV = 0x81, + SSSRAID_CMD_NONIO_FROMDEV = 0x82, +}; + +enum { + SSSRAID_QUEUE_PHYS_CONTIG = (1 << 0), + SSSRAID_CQ_IRQ_ENABLED = (1 << 1), + + SSSRAID_FEAT_NUM_QUEUES = 0x07, + SSSRAID_FEAT_ASYNC_EVENT = 0x0b, + SSSRAID_FEAT_TIMESTAMP = 0x0e, +}; + +enum sssraid_state { + SSSRAID_NEW, + SSSRAID_LIVE, + SSSRAID_RESETTING, + SSSRAID_DELETING, + SSSRAID_DEAD, +}; + +enum { + SSSRAID_CARD_HBA, + SSSRAID_CARD_RAID, +}; + +enum sssraid_cmd_type { + SSSRAID_CMD_ADM, + SSSRAID_CMD_IOPT, +}; + +/* + * SSSRAID completion queue entry struct + */ +struct sssraid_completion { + __le32 result; + union { + struct { + __u8 sense_len; + __u8 resv[3]; + }; + __le32 result1; + }; + __le16 sq_head; + __le16 sq_id; + __le16 cmd_id; + __le16 status; +}; + +/* + * SSSRAID firmware controller properties + */ +struct sssraid_ctrl_info { + __le32 nd; + __le16 max_cmds; + __le16 max_channel; + __le32 max_tgt_id; + __le16 max_lun; + __le16 max_num_sge; + __le16 lun_num_in_boot; + __u8 mdts; + __u8 acl; + __u8 aerl; + __u8 card_type; + __u16 rsvd; + __le32 rtd3e; + __u8 sn[32]; + __u8 fr[16]; + __u8 rsvd1[4020]; +}; + +struct sssraid_intr_info { + struct sssraid_dev *hdev; + u16 msix_index; + struct sssraid_cqueue *cqinfo; + char name[SSSRAID_NAME_LENGTH]; +}; + +struct sssraid_fwevt { + struct list_head list; + struct work_struct work; + struct sssraid_dev *hdev; + u16 event_id; + bool send_ack; + bool process_evt; + u32 evt_ctx; + struct kref ref_count; + char event_data[0] __aligned(4); +}; + +/* + * SSSRAID private device struct definition + */ +struct sssraid_dev { + struct pci_dev *pdev; + struct Scsi_Host *shost; + struct sssraid_squeue *sqinfo; + struct sssraid_cqueue *cqinfo; + struct dma_pool *prp_page_pool; + struct dma_pool *prp_small_pool[MAX_SMALL_POOL_NUM]; + void __iomem *bar; + + u32 init_done_queue_cnt; + u32 ioq_depth; + u32 db_stride; + u32 __iomem *dbs; + struct rw_semaphore devices_rwsem; + int numa_node; + u32 page_size; + u32 ctrl_config; + u64 cap; + u32 instance; + u32 scsi_qd; + struct sssraid_ctrl_info *ctrl_info; + struct sssraid_dev_info *devices; + + int logging_level; + + char name[SSSRAID_NAME_LENGTH]; + int cpu_count; + /* + * before_affinity_msix_cnt is + * min("FW support IO Queue count", num_online_cpus)+1 + */ + u16 before_affinity_msix_cnt; + + struct sssraid_cmd *adm_cmds; + struct list_head adm_cmd_list; + spinlock_t adm_cmd_lock; + + struct sssraid_cmd *ioq_ptcmds; + struct list_head ioq_pt_list; + spinlock_t ioq_pt_lock; + + int reset_flag; + + enum sssraid_state state; + spinlock_t state_lock; + + struct request_queue *bsg_queue; + + u8 intr_enabled; + + struct sssraid_intr_info *intr_info; + u32 intr_info_count; + + char watchdog_work_q_name[20]; + struct workqueue_struct *watchdog_work_q; + struct delayed_work watchdog_work; + spinlock_t watchdog_lock; + + char fwevt_worker_name[SSSRAID_NAME_LENGTH]; + struct workqueue_struct *fwevt_worker_thread; + spinlock_t fwevt_lock; + struct list_head fwevt_list; + + struct sssraid_fwevt *current_event; +}; + +/* + * SSSRAID scatter list descriptor + */ +struct sssraid_sgl_desc { + __le64 addr; + __le32 length; + __u8 rsvd[3]; + __u8 type; +}; + +union sssraid_data_ptr { + struct { + __le64 prp1; + __le64 prp2; + }; + struct sssraid_sgl_desc sgl; +}; + +/* + * SSSRAID general admin class command format struct + */ +struct sssraid_admin_common_command { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __le32 cdw2[4]; + union sssraid_data_ptr dptr; + __le32 cdw10; + __le32 cdw11; + __le32 cdw12; + __le32 cdw13; + __le32 cdw14; + __le32 cdw15; +}; + +struct sssraid_features { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __u64 rsvd2[2]; + union sssraid_data_ptr dptr; + __le32 fid; + __le32 dword11; + __le32 dword12; + __le32 dword13; + __le32 dword14; + __le32 dword15; +}; + +/* + * SSSRAID create completion queue command struct + */ +struct sssraid_create_cq { + __u8 opcode; + __u8 flags; + __le16 command_id; + __u32 rsvd1[5]; + __le64 prp1; + __u64 rsvd8; + __le16 cqid; + __le16 qsize; + __le16 cq_flags; + __le16 irq_vector; + __u32 rsvd12[4]; +}; + +/* + * SSSRAID create submission queue command struct + */ +struct sssraid_create_sq { + __u8 opcode; + __u8 flags; + __le16 command_id; + __u32 rsvd1[5]; + __le64 prp1; + __u64 rsvd8; + __le16 sqid; + __le16 qsize; + __le16 sq_flags; + __le16 cqid; + __u32 rsvd12[4]; +}; + +/* + * SSSRAID delete submission queue command struct + */ +struct sssraid_delete_queue { + __u8 opcode; + __u8 flags; + __le16 command_id; + __u32 rsvd1[9]; + __le16 qid; + __u16 rsvd10; + __u32 rsvd11[5]; +}; + +/* + * SSSRAID access to information command struct + */ +struct sssraid_get_info { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __u32 rsvd2[4]; + union sssraid_data_ptr dptr; + __u8 type; + __u8 rsvd10[3]; + __le32 cdw11; + __u32 rsvd12[4]; +}; + +/* + * User command struct + */ +struct sssraid_usr_cmd { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + union { + struct { + __le16 subopcode; + __le16 rsvd1; + } info_0; + __le32 cdw2; + }; + union { + struct { + __le16 data_len; + __le16 param_len; + } info_1; + __le32 cdw3; + }; + __u64 metadata; + union sssraid_data_ptr dptr; + __le32 cdw10; + __le32 cdw11; + __le32 cdw12; + __le32 cdw13; + __le32 cdw14; + __le32 cdw15; +}; + +enum { + SSSRAID_CMD_FLAG_SGL_METABUF = (1 << 6), + SSSRAID_CMD_FLAG_SGL_METASEG = (1 << 7), + SSSRAID_CMD_FLAG_SGL_ALL = SSSRAID_CMD_FLAG_SGL_METABUF | SSSRAID_CMD_FLAG_SGL_METASEG, +}; + +enum sssraid_cmd_state { + SSSRAID_CMD_IDLE = 0, + SSSRAID_CMD_IN_FLIGHT = 1, + SSSRAID_CMD_COMPLETE = 2, + SSSRAID_CMD_TIMEOUT = 3, + SSSRAID_CMD_TMO_COMPLETE = 4, +}; + +/* + * SSSRAID abort command struct + */ +struct sssraid_abort_cmd { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __u64 rsvd2[4]; + __le16 sqid; + __le16 cid; + __u32 rsvd11[5]; +}; + +/* + * SSSRAID reset command struct + */ +struct sssraid_reset_cmd { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __u64 rsvd2[4]; + __u8 type; + __u8 rsvd10[3]; + __u32 rsvd11[5]; +}; + +/* + * SSSRAID admin class command set struct + */ +struct sssraid_admin_command { + union { + struct sssraid_admin_common_command common; + struct sssraid_features features; + struct sssraid_create_cq create_cq; + struct sssraid_create_sq create_sq; + struct sssraid_delete_queue delete_queue; + struct sssraid_get_info get_info; + struct sssraid_abort_cmd abort; + struct sssraid_reset_cmd reset; + struct sssraid_usr_cmd usr_cmd; + }; +}; + +/* + * SSSRAID general IO class command format struct + */ +struct sssraid_ioq_common_command { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __le16 sense_len; + __u8 cdb_len; + __u8 rsvd2; + __le32 cdw3[3]; + union sssraid_data_ptr dptr; + __le32 cdw10[6]; + __u8 cdb[32]; + __le64 sense_addr; + __le32 cdw26[6]; +}; + +/* + * SSSRAID read or write command struct + */ +struct sssraid_rw_command { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __le16 sense_len; + __u8 cdb_len; + __u8 rsvd2; + __u32 rsvd3[3]; + union sssraid_data_ptr dptr; + __le64 slba; + __le16 nlb; + __le16 control; + __u32 rsvd13[3]; + __u8 cdb[32]; + __le64 sense_addr; + __u32 rsvd26[6]; +}; + +struct sssraid_scsi_nonio { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __le16 sense_len; + __u8 cdb_length; + __u8 rsvd2; + __u32 rsvd3[3]; + union sssraid_data_ptr dptr; + __u32 rsvd10[5]; + __le32 buffer_len; + __u8 cdb[32]; + __le64 sense_addr; + __u32 rsvd26[6]; +}; + +/* + * SSSRAID IO class command struct + */ +struct sssraid_ioq_command { + union { + struct sssraid_ioq_common_command common; + struct sssraid_rw_command rw; + struct sssraid_scsi_nonio scsi_nonio; + }; +}; + +/* + * SSSRAID passthru command struct + */ +struct sssraid_passthru_common_cmd { + __u8 opcode; + __u8 flags; + __u16 rsvd0; + __u32 nsid; + union { + struct { + __u16 subopcode; + __u16 rsvd1; + } info_0; + __u32 cdw2; + }; + union { + struct { + __u16 data_len; + __u16 param_len; + } info_1; + __u32 cdw3; + }; + __u64 metadata; + + __u64 addr; + __u64 prp2; + + __u32 cdw10; + __u32 cdw11; + __u32 cdw12; + __u32 cdw13; + __u32 cdw14; + __u32 cdw15; + __u32 timeout_ms; + __u32 result0; + __u32 result1; +}; + +struct sssraid_ioq_passthru_cmd { + __u8 opcode; + __u8 flags; + __u16 rsvd0; + __u32 nsid; + union { + struct { + __u16 res_sense_len; + __u8 cdb_len; + __u8 rsvd0; + } info_0; + __u32 cdw2; + }; + union { + struct { + __u16 subopcode; + __u16 rsvd1; + } info_1; + __u32 cdw3; + }; + union { + struct { + __u16 rsvd; + __u16 param_len; + } info_2; + __u32 cdw4; + }; + __u32 cdw5; + __u64 addr; + __u64 prp2; + union { + struct { + __u16 eid; + __u16 sid; + } info_3; + __u32 cdw10; + }; + union { + struct { + __u16 did; + __u8 did_flag; + __u8 rsvd2; + } info_4; + __u32 cdw11; + }; + __u32 cdw12; + __u32 cdw13; + __u32 cdw14; + __u32 data_len; + __u32 cdw16; + __u32 cdw17; + __u32 cdw18; + __u32 cdw19; + __u32 cdw20; + __u32 cdw21; + __u32 cdw22; + __u32 cdw23; + __u64 sense_addr; + __u32 cdw26[4]; + __u32 timeout_ms; + __u32 result0; + __u32 result1; +}; + +struct sssraid_bsg_request { + u32 msgcode; + u32 control; + union { + struct sssraid_passthru_common_cmd admcmd; + struct sssraid_ioq_passthru_cmd ioqcmd; + }; +}; + +enum { + SSSRAID_BSG_ADM, + SSSRAID_BSG_IOQ, +}; + +/* + * define the transfer command struct + */ +struct sssraid_cmd { + u16 qid; + u16 cid; + u32 result0; + u32 result1; + u16 status; + void *priv; + enum sssraid_cmd_state state; + struct completion cmd_done; + struct list_head list; +}; + +/* + * define the SSSRAID physical queue struct + */ +struct sssraid_squeue { + struct sssraid_dev *hdev; + spinlock_t sq_lock; /* spinlock for lock handling */ + + void *sq_cmds; + + dma_addr_t sq_dma_addr; + u32 __iomem *q_db; + u8 cq_phase; + u8 sqes; + u16 qidx; + u16 sq_tail; + u16 last_cq_head; + u16 q_depth; + void *sense; + dma_addr_t sense_dma_addr; + struct dma_pool *prp_small_pool; +}; + +struct sssraid_cqueue { + struct sssraid_dev *hdev; + + spinlock_t cq_lock ____cacheline_aligned_in_smp; /* spinlock for lock handling */ + + struct sssraid_completion *cqes; + + dma_addr_t cq_dma_addr; + u8 cq_phase; + u16 cq_head; + u16 last_cq_head; +}; + +/* + * define the SSSRAID IO queue descriptor struct + */ +struct sssraid_iod { + struct sssraid_squeue *sqinfo; + enum sssraid_cmd_state state; + int npages; + u32 nsge; + u32 length; + bool use_sgl; + dma_addr_t first_dma; + void *sense; + dma_addr_t sense_dma; + struct scatterlist *sg; + void *list[0]; +}; + +/* + * define the SSSRAID scsi device attribution and information + */ +#define SSSRAID_DEV_INFO_ATTR_BOOT(attr) ((attr) & 0x01) +#define SSSRAID_DEV_INFO_ATTR_VD(attr) (((attr) & 0x02) == 0x0) +#define SSSRAID_DEV_INFO_ATTR_PT(attr) (((attr) & 0x22) == 0x02) +#define SSSRAID_DEV_INFO_ATTR_RAWDISK(attr) ((attr) & 0x20) +#define SSSRAID_DEV_DISK_TYPE(attr) ((attr) & 0x1e) + +#define SSSRAID_DEV_INFO_FLAG_VALID(flag) ((flag) & 0x01) +#define SSSRAID_DEV_INFO_FLAG_CHANGE(flag) ((flag) & 0x02) + +/* + * define the SSSRAID scsi device identifier + */ +enum { + SSSRAID_SAS_HDD_VD = 0x04, + SSSRAID_SATA_HDD_VD = 0x08, + SSSRAID_SAS_SSD_VD = 0x0c, + SSSRAID_SATA_SSD_VD = 0x10, + SSSRAID_NVME_SSD_VD = 0x14, + SSSRAID_SAS_HDD_PD = 0x06, + SSSRAID_SATA_HDD_PD = 0x0a, + SSSRAID_SAS_SSD_PD = 0x0e, + SSSRAID_SATA_SSD_PD = 0x12, + SSSRAID_NVME_SSD_PD = 0x16, +}; + +/* + * define the SSSRAID scsi device queue depth + */ +#define SSSRAID_HDD_PD_QD 64 +#define SSSRAID_HDD_VD_QD 256 +#define SSSRAID_SSD_PD_QD 64 +#define SSSRAID_SSD_VD_QD 256 + +#define BGTASK_TYPE_REBUILD 4 +#define USR_CMD_READ 0xc2 +#define USR_CMD_RDLEN 0x1000 +#define USR_CMD_VDINFO 0x704 +#define USR_CMD_BGTASK 0x504 +#define VDINFO_PARAM_LEN 0x04 + +/* + * SSSRAID virtual device information struct + */ +struct sssraid_vd_info { + __u8 name[32]; + __le16 id; + __u8 rg_id; + __u8 rg_level; + __u8 sg_num; + __u8 sg_disk_num; + __u8 vd_status; + __u8 vd_type; + __u8 rsvd1[4056]; +}; + +#define MAX_REALTIME_BGTASK_NUM 32 + +struct bgtask_info { + __u8 type; + __u8 progress; + __u8 rate; + __u8 rsvd0; + __le16 vd_id; + __le16 time_left; + __u8 rsvd1[4]; +}; + +struct sssraid_bgtask { + __u8 sw; + __u8 task_num; + __u8 rsvd[6]; + struct bgtask_info bgtask[MAX_REALTIME_BGTASK_NUM]; +}; + +/* + * SSSRAID scsi device information struct + */ +struct sssraid_dev_info { + __le32 hdid; + __le16 target; + __u8 channel; + __u8 lun; + __u8 attr; + __u8 flag; + __le16 max_io_kb; +}; + +#define IOQ_PT_DATA_LEN 4096 +#define MAX_DEV_ENTRY_PER_PAGE_4K 340 +struct sssraid_dev_list { + __le32 dev_num; + __u32 rsvd0[3]; + struct sssraid_dev_info devices[MAX_DEV_ENTRY_PER_PAGE_4K]; +}; + +/* + * SSSRAID scsi device host data struct + */ +struct sssraid_sdev_hostdata { + u32 hdid; + u16 max_io_kb; + u8 attr; + u8 flag; + u8 rg_id; + u8 rsvd[3]; +}; + +extern unsigned char small_pool_num; +extern u32 io_queue_depth; +irqreturn_t sssraid_isr_poll(int irq, void *privdata); +bool sssraid_poll_cq(struct sssraid_dev *hdev, u16 qidx, int cid); +void sssraid_submit_cmd(struct sssraid_squeue *sqinfo, const void *cmd); +int sssraid_get_dev_list(struct sssraid_dev *hdev, struct sssraid_dev_info *devices); +int sssraid_submit_admin_sync_cmd(struct sssraid_dev *hdev, struct sssraid_admin_command *cmd, + u32 *result0, u32 *result1, u32 timeout); +int sssraid_send_abort_cmd(struct sssraid_dev *hdev, u32 hdid, u16 qidx, u16 cid); +int sssraid_send_reset_cmd(struct sssraid_dev *hdev, u8 type, u32 hdid); +void sssraid_adm_timeout(struct sssraid_dev *hdev, struct sssraid_cmd *cmd); +int sssraid_init_ioc(struct sssraid_dev *hdev, u8 re_init); +void sssraid_cleanup_ioc(struct sssraid_dev *hdev, u8 re_init); +int sssraid_soft_reset_handler(struct sssraid_dev *hdev); +void sssraid_free_iod_res(struct sssraid_dev *hdev, struct sssraid_iod *iod); +bool sssraid_change_host_state(struct sssraid_dev *hdev, enum sssraid_state newstate); +int sssraid_configure_timestamp(struct sssraid_dev *hdev); +int sssraid_init_ctrl_info(struct sssraid_dev *hdev); +struct sssraid_cmd *sssraid_get_cmd(struct sssraid_dev *hdev, enum sssraid_cmd_type type); +void sssraid_put_cmd(struct sssraid_dev *hdev, struct sssraid_cmd *cmd, + enum sssraid_cmd_type type); +int sssraid_send_event_ack(struct sssraid_dev *hdev, u8 event, + u32 event_ctx, u16 cid); +struct sssraid_fwevt *sssraid_alloc_fwevt(int len); +void sssraid_fwevt_add_to_list(struct sssraid_dev *hdev, + struct sssraid_fwevt *fwevt); +void sssraid_cleanup_fwevt_list(struct sssraid_dev *hdev); +void sssraid_ioc_enable_intr(struct sssraid_dev *hdev); +void sssraid_ioc_disable_intr(struct sssraid_dev *hdev); +void sssraid_cleanup_resources(struct sssraid_dev *hdev); +void sssraid_complete_cqes(struct sssraid_dev *hdev, u16 qidx, u16 start, u16 end); +int sssraid_io_map_data(struct sssraid_dev *hdev, struct sssraid_iod *iod, + struct scsi_cmnd *scmd, struct sssraid_ioq_command *ioq_cmd); +void sssraid_map_status(struct sssraid_iod *iod, struct scsi_cmnd *scmd, + struct sssraid_completion *cqe); +void sssraid_scan_disk(struct sssraid_dev *hdev); +void sssraid_complete_aen(struct sssraid_dev *hdev, struct sssraid_completion *cqe); +void sssraid_back_all_io(struct sssraid_dev *hdev); + +static inline void **sssraid_iod_list(struct sssraid_iod *iod) +{ + return iod->list; +} + +#end diff --git a/drivers/scsi/sssraid/sssraid_debug.h b/drivers/scsi/sssraid/sssraid_debug.h new file mode 100644 index 000000000000..d48c57e275f7 --- /dev/null +++ b/drivers/scsi/sssraid/sssraid_debug.h @@ -0,0 +1,34 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2022 3SNIC Information Technology, Ltd */ + +/* 3SNIC RAID SSSXXX Series Linux Driver */ + +#ifndef SSSRAID_DEBUG_H_INCLUDED +#define SSSRAID_DEBUG_H_INCLUDED + +/* + * debug levels + */ +#define SSSRAID_DEBUG 0x00000001 + +/* + * debug macros + */ + +#define ioc_err(ioc, fmt, ...) \ + pr_err("%s: " fmt, (ioc)->name, ##__VA_ARGS__) +#define ioc_notice(ioc, fmt, ...) \ + pr_notice("%s: " fmt, (ioc)->name, ##__VA_ARGS__) +#define ioc_warn(ioc, fmt, ...) \ + pr_warn("%s: " fmt, (ioc)->name, ##__VA_ARGS__) +#define ioc_info(ioc, fmt, ...) \ + pr_info("%s: " fmt, (ioc)->name, ##__VA_ARGS__) + + +#define dbgprint(IOC, FMT, ...) \ + do { \ + if (unlikely(IOC->logging_level & SSSRAID_DEBUG)) \ + pr_info("%s: " FMT, (IOC)->name, ##__VA_ARGS__); \ + } while (0) + +#endif /* SSSRAID_DEBUG_H_INCLUDED */ diff --git a/drivers/scsi/sssraid/sssraid_fw.c b/drivers/scsi/sssraid/sssraid_fw.c new file mode 100644 index 000000000000..54287cf70558 --- /dev/null +++ b/drivers/scsi/sssraid/sssraid_fw.c @@ -0,0 +1,1724 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2022 3SNIC Information Technology, Ltd */ + +/* 3SNIC RAID SSSXXX Series Linux Driver */ + +#include <linux/version.h> +#include <linux/pci.h> +#include <linux/aer.h> +#include <linux/module.h> +#include <linux/ioport.h> +#include <linux/device.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/cdev.h> +#include <linux/sysfs.h> +#include <linux/gfp.h> +#include <linux/types.h> +#include <linux/ratelimit.h> +#include <linux/debugfs.h> +#include <linux/blkdev.h> +#include <linux/bsg-lib.h> +#include <linux/sort.h> + +#include <scsi/scsi.h> +#include <scsi/scsi_cmnd.h> +#include <scsi/scsi_device.h> +#include <scsi/scsi_host.h> +#include <scsi/scsi_transport.h> +#include <scsi/scsi_dbg.h> + +#include <linux/unaligned/be_byteshift.h> +#include <linux/unaligned/le_byteshift.h> +#include <linux/once.h> +#include <linux/sched/signal.h> +#include <linux/io-64-nonatomic-lo-hi.h> + +#include "sssraid.h" +#include "sssraid_debug.h" + +static int sssraid_wait_ready(struct sssraid_dev *hdev, u64 cap, bool enabled) +{ + unsigned long timeout = + ((SSSRAID_CAP_TIMEOUT(cap) + 1) * SSSRAID_CAP_TIMEOUT_UNIT_MS) + jiffies; + u32 bit = enabled ? SSSRAID_CSTS_RDY : 0; + + while ((readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_RDY) != bit) { + usleep_range(1000, 2000); + if (fatal_signal_pending(current)) + return -EINTR; + + if (time_after(jiffies, timeout)) { + ioc_err(hdev, "Device not ready; aborting %s\n", + enabled ? "initialisation" : "reset"); + return -ENODEV; + } + } + return 0; +} + +static int sssraid_enable_ctrl(struct sssraid_dev *hdev) +{ + u64 cap = hdev->cap; + u32 dev_page_min = SSSRAID_CAP_MPSMIN(cap) + 12; + u32 page_shift = PAGE_SHIFT; + + if (page_shift < dev_page_min) { + ioc_err(hdev, "Minimum device page size[%u], too large for host[%u]\n", + 1U << dev_page_min, 1U << page_shift); + return -ENODEV; + } + + page_shift = min_t(unsigned int, SSSRAID_CAP_MPSMAX(cap) + 12, PAGE_SHIFT); + hdev->page_size = 1U << page_shift; + + hdev->ctrl_config = SSSRAID_CC_CSS_NVM; + hdev->ctrl_config |= (page_shift - 12) << SSSRAID_CC_MPS_SHIFT; + hdev->ctrl_config |= SSSRAID_CC_AMS_RR | SSSRAID_CC_SHN_NONE; + hdev->ctrl_config |= SSSRAID_CC_IOSQES | SSSRAID_CC_IOCQES; + hdev->ctrl_config |= SSSRAID_CC_ENABLE; + writel(hdev->ctrl_config, hdev->bar + SSSRAID_REG_CC); + + return sssraid_wait_ready(hdev, cap, true); +} + +static int sssraid_disable_ctrl(struct sssraid_dev *hdev) +{ + hdev->ctrl_config &= ~SSSRAID_CC_SHN_MASK; + hdev->ctrl_config &= ~SSSRAID_CC_ENABLE; + writel(hdev->ctrl_config, hdev->bar + SSSRAID_REG_CC); + + return sssraid_wait_ready(hdev, hdev->cap, false); +} + +static int sssraid_shutdown_ctrl(struct sssraid_dev *hdev) +{ + unsigned long timeout = le32_to_cpu(hdev->ctrl_info->rtd3e) / 1000000 * HZ + jiffies; + + hdev->ctrl_config &= ~SSSRAID_CC_SHN_MASK; + hdev->ctrl_config |= SSSRAID_CC_SHN_NORMAL; + writel(hdev->ctrl_config, hdev->bar + SSSRAID_REG_CC); + + while ((readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_SHST_MASK) != + SSSRAID_CSTS_SHST_CMPLT) { + msleep(100); + if (fatal_signal_pending(current)) + return -EINTR; + if (time_after(jiffies, timeout)) { + ioc_err(hdev, "Device shutdown incomplete; abort shutdown\n"); + return -ENODEV; + } + } + return 0; +} + +static int sssraid_remap_bar(struct sssraid_dev *hdev, u32 size) +{ + struct pci_dev *pdev = hdev->pdev; + + if (size > pci_resource_len(pdev, 0)) { + ioc_err(hdev, "Input size[%u] exceed bar0 length[%llu]\n", + size, pci_resource_len(pdev, 0)); + return -ENODEV; + } + + if (hdev->bar) + iounmap(hdev->bar); + + hdev->bar = ioremap(pci_resource_start(pdev, 0), size); + if (!hdev->bar) { + ioc_err(hdev, "ioremap for bar0 failed\n"); + return -ENODEV; + } + hdev->dbs = hdev->bar + SSSRAID_REG_DBS; + + return 0; +} + +static int sssraid_create_dma_pools(struct sssraid_dev *hdev) +{ + int i; + char poolname[20] = { 0 }; + + hdev->prp_page_pool = dma_pool_create("prp list page", &hdev->pdev->dev, + PAGE_SIZE, PAGE_SIZE, 0); + + if (!hdev->prp_page_pool) { + ioc_err(hdev, "create prp_page_pool failed\n"); + return -ENOMEM; + } + + for (i = 0; i < small_pool_num; i++) { + sprintf(poolname, "prp_list_256_%d", i); + hdev->prp_small_pool[i] = dma_pool_create(poolname, &hdev->pdev->dev, + SMALL_POOL_SIZE, + SMALL_POOL_SIZE, 0); + + if (!hdev->prp_small_pool[i]) { + ioc_err(hdev, "create prp_small_pool %d failed\n", i); + goto destroy_prp_small_pool; + } + } + + return 0; + +destroy_prp_small_pool: + while (i > 0) + dma_pool_destroy(hdev->prp_small_pool[--i]); + dma_pool_destroy(hdev->prp_page_pool); + + return -ENOMEM; +} + +static void sssraid_destroy_dma_pools(struct sssraid_dev *hdev) +{ + int i; + + for (i = 0; i < small_pool_num; i++) + dma_pool_destroy(hdev->prp_small_pool[i]); + dma_pool_destroy(hdev->prp_page_pool); +} + +static int sssraid_alloc_resources(struct sssraid_dev *hdev) +{ + int retval, nqueue; + + hdev->ctrl_info = kzalloc_node(sizeof(*hdev->ctrl_info), GFP_KERNEL, hdev->numa_node); + if (!hdev->ctrl_info) + return -ENOMEM; + + retval = sssraid_create_dma_pools(hdev); + if (retval) + goto free_ctrl_info; + /* not num_online_cpus */ + nqueue = num_possible_cpus() + 1; + hdev->cqinfo = kcalloc_node(nqueue, sizeof(struct sssraid_cqueue), + GFP_KERNEL, hdev->numa_node); + if (!hdev->cqinfo) { + retval = -ENOMEM; + goto destroy_dma_pools; + } + + hdev->sqinfo = kcalloc_node(nqueue, sizeof(struct sssraid_squeue), + GFP_KERNEL, hdev->numa_node); + if (!hdev->sqinfo) { + retval = -ENOMEM; + goto free_cqueues; + } + + /* sssraid_alloc_admin_cmds moved to sssraid_init_ioc */ + + ioc_info(hdev, "queues num: %d\n", nqueue); + + return 0; + +free_cqueues: + kfree(hdev->cqinfo); +destroy_dma_pools: + sssraid_destroy_dma_pools(hdev); +free_ctrl_info: + kfree(hdev->ctrl_info); + + return retval; +} + +void sssraid_ioc_enable_intr(struct sssraid_dev *hdev) +{ + hdev->intr_enabled = 1; +} + +void sssraid_ioc_disable_intr(struct sssraid_dev *hdev) +{ + u16 i, max_vectors; + + hdev->intr_enabled = 0; + max_vectors = hdev->intr_info_count; + + for (i = 0; i < max_vectors; i++) + synchronize_irq(pci_irq_vector(hdev->pdev, i)); +} + +static int sssraid_setup_resources(struct sssraid_dev *hdev) +{ + struct pci_dev *pdev = hdev->pdev; + int retval = 0; + u64 maskbit = SSSRAID_DMA_MSK_BIT_MAX; + + if (pci_enable_device_mem(pdev)) { + ioc_err(hdev, "pci_enable_device_mem: failed\n"); + retval = -ENODEV; + goto out_failed; + } + + retval = pci_request_mem_regions(pdev, SSSRAID_DRIVER_NAME); + if (retval) { + ioc_err(hdev, "fail to request memory regions\n"); + retval = -ENODEV; + goto out_failed; + } + + /* get cap value at first, so keep + * sssraid_remap_bar(hdev, SSSRAID_REG_DBS + 4096) + * ioremap(pci_resource_start(..)) still in sssraid_remap_bar + */ + retval = sssraid_remap_bar(hdev, SSSRAID_REG_DBS + 4096); + if (retval) { + retval = -ENODEV; + goto out_failed; + } + + pci_set_master(pdev); + + if (readl(hdev->bar + SSSRAID_REG_CSTS) == U32_MAX) { + retval = -ENODEV; + ioc_err(hdev, "Read csts register failed\n"); + goto out_failed; + } + + hdev->cap = lo_hi_readq(hdev->bar + SSSRAID_REG_CAP); + hdev->ioq_depth = min_t(u32, SSSRAID_CAP_MQES(hdev->cap) + 1, io_queue_depth); + hdev->scsi_qd = hdev->ioq_depth - SSSRAID_PTCMDS_PERQ; + hdev->db_stride = 1 << SSSRAID_CAP_STRIDE(hdev->cap); + + maskbit = SSSRAID_CAP_DMAMASK(hdev->cap); + if (maskbit < 32 || maskbit > SSSRAID_DMA_MSK_BIT_MAX) { + ioc_err(hdev, "err, dma mask invalid[%llu], set to default\n", maskbit); + maskbit = SSSRAID_DMA_MSK_BIT_MAX; + } + + if (dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(maskbit))) { + if (dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32))) { + ioc_err(hdev, "set dma mask[32] and coherent failed\n"); + retval = -ENODEV; + goto out_failed; + } + ioc_info(hdev, "set dma mask[32] success\n"); + } else { + ioc_info(hdev, "set dma mask[%llu] success\n", maskbit); + } + + /* pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_ALL_TYPES) moved to setup_isr */ + + pci_set_drvdata(pdev, hdev->shost); + + pci_enable_pcie_error_reporting(pdev); + pci_save_state(pdev); + + sssraid_ioc_disable_intr(hdev); + + return retval; + +out_failed: + sssraid_cleanup_resources(hdev); + return retval; +} + +static int sssraid_alloc_admin_cmds(struct sssraid_dev *hdev) +{ + u16 i; + + INIT_LIST_HEAD(&hdev->adm_cmd_list); + spin_lock_init(&hdev->adm_cmd_lock); + + hdev->adm_cmds = kcalloc_node(SSSRAID_AQ_BLK_MQ_DEPTH, sizeof(struct sssraid_cmd), + GFP_KERNEL, hdev->numa_node); + + if (!hdev->adm_cmds) { + ioc_err(hdev, "Alloc admin cmds failed\n"); + return -ENOMEM; + } + + for (i = 0; i < SSSRAID_AQ_BLK_MQ_DEPTH; i++) { + hdev->adm_cmds[i].qid = 0; + hdev->adm_cmds[i].cid = i; + list_add_tail(&(hdev->adm_cmds[i].list), &hdev->adm_cmd_list); + } + + ioc_info(hdev, "Alloc admin cmds success, num[%d]\n", SSSRAID_AQ_BLK_MQ_DEPTH); + + return 0; +} + +static int sssraid_alloc_qpair(struct sssraid_dev *hdev, u16 qidx, u16 depth) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + int retval = 0; + + if (hdev->init_done_queue_cnt > qidx) { + ioc_info(hdev, "warn: queue[%d] is exist\n", qidx); + return 0; + } + + cqinfo->cqes = dma_alloc_coherent(&hdev->pdev->dev, CQ_SIZE(depth), + &cqinfo->cq_dma_addr, GFP_KERNEL | __GFP_ZERO); + if (!cqinfo->cqes) + return -ENOMEM; + + sqinfo->sq_cmds = dma_alloc_coherent(&hdev->pdev->dev, SQ_SIZE(qidx, depth), + &sqinfo->sq_dma_addr, GFP_KERNEL); + if (!sqinfo->sq_cmds) { + retval = -ENOMEM; + goto free_cqes; + } + + /* alloc sense buffer */ + sqinfo->sense = dma_alloc_coherent(&hdev->pdev->dev, SENSE_SIZE(depth), + &sqinfo->sense_dma_addr, GFP_KERNEL | __GFP_ZERO); + if (!sqinfo->sense) { + retval = -ENOMEM; + goto free_sq_cmds; + } + + spin_lock_init(&sqinfo->sq_lock); + spin_lock_init(&cqinfo->cq_lock); + cqinfo->hdev = hdev; + sqinfo->hdev = hdev; + sqinfo->q_depth = depth; + sqinfo->qidx = qidx; + /* cq_vector replaced by msix_index */ + + /* + * online_queues: completely initialized queue count: sssraid_init_queue + * queue_count: allocated but not completely initialized queue count: sssraid_alloc_queue + * online_queues/queue_count replaced by init_done_queue_cnt. + */ + hdev->init_done_queue_cnt++; + + return 0; + +free_sq_cmds: + dma_free_coherent(&hdev->pdev->dev, SQ_SIZE(qidx, depth), (void *)sqinfo->sq_cmds, + sqinfo->sq_dma_addr); +free_cqes: + dma_free_coherent(&hdev->pdev->dev, CQ_SIZE(depth), (void *)cqinfo->cqes, + cqinfo->cq_dma_addr); + return retval; +} + +static void sssraid_init_queue(struct sssraid_dev *hdev, u16 qidx) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + + memset((void *)cqinfo->cqes, 0, CQ_SIZE(sqinfo->q_depth)); + + sqinfo->sq_tail = 0; + cqinfo->cq_head = 0; + cqinfo->cq_phase = 1; + sqinfo->q_db = &hdev->dbs[qidx * 2 * hdev->db_stride]; + sqinfo->prp_small_pool = hdev->prp_small_pool[qidx % small_pool_num]; +} + +static int sssraid_setup_admin_qpair(struct sssraid_dev *hdev) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[0]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[0]; + u32 aqa; + int retval; + + ioc_info(hdev, "start disable ctrl\n"); + + retval = sssraid_disable_ctrl(hdev); + if (retval) + return retval; + + /* this func don't alloc admin queue */ + + aqa = sqinfo->q_depth - 1; + aqa |= aqa << 16; + writel(aqa, hdev->bar + SSSRAID_REG_AQA); + lo_hi_writeq(sqinfo->sq_dma_addr, hdev->bar + SSSRAID_REG_ASQ); + lo_hi_writeq(cqinfo->cq_dma_addr, hdev->bar + SSSRAID_REG_ACQ); + + ioc_info(hdev, "start enable ctrl\n"); + + retval = sssraid_enable_ctrl(hdev); + if (retval) { + retval = -ENODEV; + return retval; + } + + /* interrupt registry not here */ + /* cq_vector replaced by msix_index */ + + sssraid_init_queue(hdev, 0); + + ioc_info(hdev, "success, queuecount:[%d], pagesize[%d]\n", + hdev->init_done_queue_cnt, hdev->page_size); + + return 0; +} + +static void sssraid_cleanup_isr(struct sssraid_dev *hdev) +{ + u16 i; + + sssraid_ioc_disable_intr(hdev); + + if (!hdev->intr_info) + return; + + for (i = 0; i < hdev->intr_info_count; i++) + free_irq(pci_irq_vector(hdev->pdev, i), + (hdev->intr_info + i)); + + kfree(hdev->intr_info); + hdev->intr_info = NULL; + hdev->intr_info_count = 0; + pci_free_irq_vectors(hdev->pdev); +} + +static void sssraid_complete_adminq_cmnd(struct sssraid_dev *hdev, u16 qidx, + struct sssraid_completion *cqe) +{ + struct sssraid_cmd *adm_cmd; + + adm_cmd = hdev->adm_cmds + le16_to_cpu(cqe->cmd_id); + if (unlikely(adm_cmd->state == SSSRAID_CMD_IDLE)) { + ioc_warn(hdev, "Invalid id %d completed on queue %d\n", + le16_to_cpu(cqe->cmd_id), le16_to_cpu(cqe->sq_id)); + return; + } + + adm_cmd->status = le16_to_cpu(cqe->status) >> 1; + adm_cmd->result0 = le32_to_cpu(cqe->result); + adm_cmd->result1 = le32_to_cpu(cqe->result1); + + complete(&adm_cmd->cmd_done); +} + +static inline bool sssraid_cqe_pending(struct sssraid_cqueue *cqinfo) +{ + return (le16_to_cpu(cqinfo->cqes[cqinfo->cq_head].status) & 1) == + cqinfo->cq_phase; +} + +static inline void sssraid_update_cq_head(struct sssraid_dev *hdev, u16 qidx) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + + if (++cqinfo->cq_head == sqinfo->q_depth) { + cqinfo->cq_head = 0; + cqinfo->cq_phase = !cqinfo->cq_phase; + } +} + +static inline bool sssraid_process_cq(struct sssraid_dev *hdev, u16 qidx, u16 *start, + u16 *end, int tag) +{ + bool found = false; + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + + *start = cqinfo->cq_head; + while (!found && sssraid_cqe_pending(cqinfo)) { + if (le16_to_cpu(cqinfo->cqes[cqinfo->cq_head].cmd_id) == tag) + found = true; + sssraid_update_cq_head(hdev, qidx); + } + *end = cqinfo->cq_head; + + if (*start != *end) + writel(cqinfo->cq_head, sqinfo->q_db + sqinfo->hdev->db_stride); + + return found; +} + +static irqreturn_t sssraid_isr(int irq, void *privdata) +{ + struct sssraid_intr_info *intr_info = privdata; + struct sssraid_dev *hdev = intr_info->hdev; + irqreturn_t ret = IRQ_NONE; + struct sssraid_cqueue *cqinfo; + u16 midx, start, end; + + if (!intr_info) + return IRQ_NONE; + + if (!hdev->intr_enabled) + return IRQ_NONE; + + midx = intr_info->msix_index; + cqinfo = &hdev->cqinfo[midx]; + + spin_lock(&cqinfo->cq_lock); + if (cqinfo->cq_head != cqinfo->last_cq_head) + ret = IRQ_HANDLED; + + sssraid_process_cq(hdev, midx, &start, &end, -1); + cqinfo->last_cq_head = cqinfo->cq_head; + spin_unlock(&cqinfo->cq_lock); + + if (start != end) { + sssraid_complete_cqes(hdev, midx, start, end); + ret = IRQ_HANDLED; + } + return ret; +} + +irqreturn_t sssraid_isr_poll(int irq, void *privdata) +{ + return IRQ_NONE; +} + +bool sssraid_poll_cq(struct sssraid_dev *hdev, u16 qidx, int cid) +{ + u16 start, end; + bool found; + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + + if (!sssraid_cqe_pending(cqinfo)) + return 0; + + spin_lock_irq(&cqinfo->cq_lock); + found = sssraid_process_cq(hdev, qidx, &start, &end, cid); + spin_unlock_irq(&cqinfo->cq_lock); + + sssraid_complete_cqes(hdev, qidx, start, end); + return found; +} + +static inline int sssraid_request_irq(struct sssraid_dev *hdev, u16 index) +{ + struct pci_dev *pdev = hdev->pdev; + struct sssraid_intr_info *intr_info = hdev->intr_info + index; + int retval = 0; + + intr_info->hdev = hdev; + intr_info->msix_index = index; + intr_info->cqinfo = NULL; + + snprintf(intr_info->name, SSSRAID_NAME_LENGTH, "%s%d-msix%d", + SSSRAID_DRIVER_NAME, hdev->instance, index); + + retval = request_threaded_irq(pci_irq_vector(pdev, index), sssraid_isr, + sssraid_isr_poll, IRQF_SHARED, intr_info->name, intr_info); + + if (retval) { + ioc_err(hdev, "%s: Unable to allocate interrupt %d!\n", + intr_info->name, pci_irq_vector(pdev, index)); + return retval; + } + + return retval; +} + +static int sssraid_setup_isr(struct sssraid_dev *hdev, u8 setup_one) +{ + unsigned int irq_flags = PCI_IRQ_MSIX; + u16 max_vectors = 0, i; + int retval = 0; + + struct irq_affinity desc = { .pre_vectors = 1}; + + sssraid_cleanup_isr(hdev); + + if (setup_one) + max_vectors = 1; + else { + max_vectors = hdev->before_affinity_msix_cnt; + + ioc_info(hdev, "MSI-x vectors requested: %d\n", max_vectors); + } + + irq_flags |= PCI_IRQ_AFFINITY | PCI_IRQ_ALL_TYPES; + + i = pci_alloc_irq_vectors_affinity(hdev->pdev, + 1, max_vectors, irq_flags, &desc); + + if (i <= 0) { + ioc_err(hdev, "Cannot alloc irq vectors\n"); + goto out_failed; + } + if (i != max_vectors) { + ioc_info(hdev, + "allocated vectors (%d) are less than configured (%d)\n", + i, max_vectors); + + max_vectors = i; + } + + hdev->intr_info = kzalloc(sizeof(struct sssraid_intr_info) * max_vectors, + GFP_KERNEL); + if (!hdev->intr_info) { + retval = -1; + pci_free_irq_vectors(hdev->pdev); + goto out_failed; + } + + for (i = 0; i < max_vectors; i++) { + retval = sssraid_request_irq(hdev, i); + if (retval) { + hdev->intr_info_count = i; /* =i is for offload interrupt loop counter */ + goto out_failed; + } + } + + /* intr_info_count replace max_qid */ + hdev->intr_info_count = max_vectors; + sssraid_ioc_enable_intr(hdev); + return retval; +out_failed: + sssraid_cleanup_isr(hdev); + + return retval; +} + +static bool sssraid_adm_need_reset(struct sssraid_admin_command *cmd) +{ + switch (cmd->common.opcode) { + case SSSRAID_ADMIN_DELETE_SQ: + case SSSRAID_ADMIN_CREATE_SQ: + case SSSRAID_ADMIN_DELETE_CQ: + case SSSRAID_ADMIN_CREATE_CQ: + case SSSRAID_ADMIN_SET_FEATURES: + return false; + default: + return true; + } +} + +void sssraid_submit_cmd(struct sssraid_squeue *sqinfo, const void *cmd) +{ + u32 sqes = SQE_SIZE(sqinfo->qidx); + unsigned long flags; + struct sssraid_admin_common_command *acd = (struct sssraid_admin_common_command *)cmd; + + spin_lock_irqsave(&sqinfo->sq_lock, flags); + memcpy((sqinfo->sq_cmds + sqes * sqinfo->sq_tail), cmd, sqes); + if (++sqinfo->sq_tail == sqinfo->q_depth) + sqinfo->sq_tail = 0; + + writel(sqinfo->sq_tail, sqinfo->q_db); + spin_unlock_irqrestore(&sqinfo->sq_lock, flags); + + dbgprint(sqinfo->hdev, "cid[%d] qidx[%d], opcode[0x%x], flags[0x%x], hdid[%u]\n", + le16_to_cpu(acd->command_id), sqinfo->qidx, acd->opcode, acd->flags, + le32_to_cpu(acd->hdid)); +} + +int sssraid_submit_admin_sync_cmd(struct sssraid_dev *hdev, struct sssraid_admin_command *cmd, + u32 *result0, u32 *result1, u32 timeout) +{ + struct sssraid_cmd *adm_cmd = sssraid_get_cmd(hdev, SSSRAID_CMD_ADM); + + if (!adm_cmd) { + ioc_err(hdev, "err, get admin cmd failed\n"); + return -EFAULT; + } + + timeout = timeout ? timeout : ADMIN_TIMEOUT; + + /* + * watch dog not as optimized as + * init_completion/complete + */ + init_completion(&adm_cmd->cmd_done); + + cmd->common.command_id = cpu_to_le16(adm_cmd->cid); + sssraid_submit_cmd(&hdev->sqinfo[0], cmd); + + if (!wait_for_completion_timeout(&adm_cmd->cmd_done, timeout)) { + ioc_err(hdev, "cid[%d] qidx[%d] timeout, opcode[0x%x] subopcode[0x%x]\n", + adm_cmd->cid, adm_cmd->qid, cmd->usr_cmd.opcode, + cmd->usr_cmd.info_0.subopcode); + + /* reset controller if admin timeout */ + if (sssraid_adm_need_reset(cmd)) + sssraid_adm_timeout(hdev, adm_cmd); + + sssraid_put_cmd(hdev, adm_cmd, SSSRAID_CMD_ADM); + return -ETIME; + } + + if (result0) + *result0 = adm_cmd->result0; + if (result1) + *result1 = adm_cmd->result1; + + sssraid_put_cmd(hdev, adm_cmd, SSSRAID_CMD_ADM); + + return adm_cmd->status; +} + +static int sssraid_get_ctrl_info(struct sssraid_dev *hdev, struct sssraid_ctrl_info *ctrl_info) +{ + struct sssraid_admin_command admin_cmd; + u8 *data_ptr = NULL; + dma_addr_t data_dma = 0; + int retval; + + data_ptr = dma_alloc_coherent(&hdev->pdev->dev, PAGE_SIZE, &data_dma, GFP_KERNEL); + if (!data_ptr) + return -ENOMEM; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.get_info.opcode = SSSRAID_ADMIN_GET_INFO; + admin_cmd.get_info.type = SSSRAID_GET_INFO_CTRL; + admin_cmd.common.dptr.prp1 = cpu_to_le64(data_dma); + + retval = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); + if (!retval) + memcpy(ctrl_info, data_ptr, sizeof(struct sssraid_ctrl_info)); + + dma_free_coherent(&hdev->pdev->dev, PAGE_SIZE, data_ptr, data_dma); + + return retval; +} + +int sssraid_init_ctrl_info(struct sssraid_dev *hdev) +{ + int retval; + + hdev->ctrl_info->nd = cpu_to_le32(240); + hdev->ctrl_info->mdts = 8; + hdev->ctrl_info->max_cmds = cpu_to_le16(4096); + hdev->ctrl_info->max_num_sge = cpu_to_le16(128); + hdev->ctrl_info->max_channel = cpu_to_le16(4); + hdev->ctrl_info->max_tgt_id = cpu_to_le32(3239); + hdev->ctrl_info->max_lun = cpu_to_le16(2); + + retval = sssraid_get_ctrl_info(hdev, hdev->ctrl_info); + if (retval) + ioc_err(hdev, "init controller info failed: %d\n", retval); + + ioc_info(hdev, "nd = %d\n", le32_to_cpu(hdev->ctrl_info->nd)); + ioc_info(hdev, "max_cmd = %d\n", le16_to_cpu(hdev->ctrl_info->max_cmds)); + ioc_info(hdev, "max_channel = %d\n", le16_to_cpu(hdev->ctrl_info->max_channel)); + ioc_info(hdev, "max_tgt_id = %d\n", le32_to_cpu(hdev->ctrl_info->max_tgt_id)); + ioc_info(hdev, "max_lun = %d\n", le16_to_cpu(hdev->ctrl_info->max_lun)); + ioc_info(hdev, "max_num_sge = %d\n", le16_to_cpu(hdev->ctrl_info->max_num_sge)); + ioc_info(hdev, "lun_num_boot = %d\n", le16_to_cpu(hdev->ctrl_info->lun_num_in_boot)); + ioc_info(hdev, "mdts = %d\n", hdev->ctrl_info->mdts); + ioc_info(hdev, "acl = %d\n", hdev->ctrl_info->acl); + ioc_info(hdev, "aer1 = %d\n", hdev->ctrl_info->aerl); + ioc_info(hdev, "card_type = %d\n", hdev->ctrl_info->card_type); + ioc_info(hdev, "rtd3e = %d\n", le32_to_cpu(hdev->ctrl_info->rtd3e)); + ioc_info(hdev, "sn = %s\n", hdev->ctrl_info->sn); + ioc_info(hdev, "fr = %s\n", hdev->ctrl_info->fr); + + if (!hdev->ctrl_info->aerl) + hdev->ctrl_info->aerl = 1; + if (hdev->ctrl_info->aerl > SSSRAID_NR_AEN_COMMANDS) + hdev->ctrl_info->aerl = SSSRAID_NR_AEN_COMMANDS; + + return 0; +} + +static int sssraid_set_features(struct sssraid_dev *hdev, u32 fid, u32 dword11, void *buffer, + size_t buflen, u32 *result) +{ + struct sssraid_admin_command admin_cmd; + int ret; + u8 *data_ptr = NULL; + dma_addr_t data_dma = 0; + + if (buffer && buflen) { + data_ptr = dma_alloc_coherent(&hdev->pdev->dev, buflen, &data_dma, GFP_KERNEL); + if (!data_ptr) + return -ENOMEM; + + memcpy(data_ptr, buffer, buflen); + } + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.features.opcode = SSSRAID_ADMIN_SET_FEATURES; + admin_cmd.features.fid = cpu_to_le32(fid); + admin_cmd.features.dword11 = cpu_to_le32(dword11); + admin_cmd.common.dptr.prp1 = cpu_to_le64(data_dma); + + ret = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, result, NULL, 0); + + if (data_ptr) + dma_free_coherent(&hdev->pdev->dev, buflen, data_ptr, data_dma); + + return ret; +} + +static int sssraid_set_queue_cnt(struct sssraid_dev *hdev, u32 *cnt) +{ + u32 q_cnt = (*cnt - 1) | ((*cnt - 1) << 16); + u32 nr_ioqs, result; + int status; + + status = sssraid_set_features(hdev, SSSRAID_FEAT_NUM_QUEUES, q_cnt, NULL, 0, &result); + if (status) { + ioc_err(hdev, "Set queue count failed, status: %d\n", + status); + return -EIO; + } + + nr_ioqs = min(result & 0xffff, result >> 16) + 1; + *cnt = min(*cnt, nr_ioqs); + if (*cnt == 0) { + ioc_err(hdev, "Illegal queue count: zero\n"); + return -EIO; + } + return 0; +} + +static int sssraid_create_cq(struct sssraid_dev *hdev, u16 qidx) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + struct sssraid_admin_command admin_cmd; + int flags = SSSRAID_QUEUE_PHYS_CONTIG | SSSRAID_CQ_IRQ_ENABLED; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.create_cq.opcode = SSSRAID_ADMIN_CREATE_CQ; + admin_cmd.create_cq.prp1 = cpu_to_le64(cqinfo->cq_dma_addr); + admin_cmd.create_cq.cqid = cpu_to_le16(qidx); + admin_cmd.create_cq.qsize = cpu_to_le16(sqinfo->q_depth - 1); + admin_cmd.create_cq.cq_flags = cpu_to_le16(flags); + admin_cmd.create_cq.irq_vector = cpu_to_le16(qidx); + + return sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); +} + +static int sssraid_create_io_cq(struct sssraid_dev *hdev, u16 qidx) +{ + int retval; + struct sssraid_cqueue *cqinfo = hdev->cqinfo + qidx; + u16 midx = qidx; + + retval = sssraid_create_cq(hdev, qidx); + if (retval) + return retval; + + /* + * cqinfo initialization at sssraid_init_queue + */ + hdev->intr_info[midx].cqinfo = cqinfo; + + return retval; +} + +static int sssraid_create_sq(struct sssraid_dev *hdev, u16 qidx) +{ + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + struct sssraid_admin_command admin_cmd; + int flags = SSSRAID_QUEUE_PHYS_CONTIG; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.create_sq.opcode = SSSRAID_ADMIN_CREATE_SQ; + admin_cmd.create_sq.prp1 = cpu_to_le64(sqinfo->sq_dma_addr); + admin_cmd.create_sq.sqid = cpu_to_le16(qidx); + admin_cmd.create_sq.qsize = cpu_to_le16(sqinfo->q_depth - 1); + admin_cmd.create_sq.sq_flags = cpu_to_le16(flags); + admin_cmd.create_sq.cqid = cpu_to_le16(qidx); + + return sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); +} + +static int sssraid_create_io_sq(struct sssraid_dev *hdev, u16 qidx) +{ + return sssraid_create_sq(hdev, qidx); +} + +int sssraid_get_dev_list(struct sssraid_dev *hdev, struct sssraid_dev_info *devices) +{ + u32 nd = le32_to_cpu(hdev->ctrl_info->nd); + struct sssraid_admin_command admin_cmd; + struct sssraid_dev_list *list_buf; + dma_addr_t data_dma = 0; + u32 i, idx, hdid, ndev; + int ret = 0; + + list_buf = dma_alloc_coherent(&hdev->pdev->dev, PAGE_SIZE, &data_dma, GFP_KERNEL); + if (!list_buf) + return -ENOMEM; + + for (idx = 0; idx < nd;) { + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.get_info.opcode = SSSRAID_ADMIN_GET_INFO; + admin_cmd.get_info.type = SSSRAID_GET_INFO_DEV_LIST; + admin_cmd.get_info.cdw11 = cpu_to_le32(idx); + admin_cmd.common.dptr.prp1 = cpu_to_le64(data_dma); + + ret = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); + + if (ret) { + ioc_err(hdev, "Get device list failed, nd: %u, idx: %u, ret: %d\n", + nd, idx, ret); + goto out; + } + ndev = le32_to_cpu(list_buf->dev_num); + + ioc_info(hdev, "ndev numbers: %u\n", ndev); + + for (i = 0; i < ndev; i++) { + hdid = le32_to_cpu(list_buf->devices[i].hdid); + ioc_info(hdev, "list_buf->devices[%d], hdid: %u target: %d, channel: %d, lun: %d, attr[0x%x]\n", + i, hdid, le16_to_cpu(list_buf->devices[i].target), + list_buf->devices[i].channel, + list_buf->devices[i].lun, + list_buf->devices[i].attr); + if (hdid > nd || hdid == 0) { + ioc_err(hdev, "err, hdid[%d] invalid\n", hdid); + continue; + } + memcpy(&devices[hdid - 1], &list_buf->devices[i], + sizeof(struct sssraid_dev_info)); + } + idx += ndev; + + if (ndev < MAX_DEV_ENTRY_PER_PAGE_4K) + break; + } + +out: + dma_free_coherent(&hdev->pdev->dev, PAGE_SIZE, list_buf, data_dma); + return ret; +} + +/* send abort command by admin queue temporary */ +int sssraid_send_abort_cmd(struct sssraid_dev *hdev, u32 hdid, u16 qidx, u16 cid) +{ + struct sssraid_admin_command admin_cmd; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.abort.opcode = SSSRAID_ADMIN_ABORT_CMD; + admin_cmd.abort.hdid = cpu_to_le32(hdid); + admin_cmd.abort.sqid = cpu_to_le16(qidx); + admin_cmd.abort.cid = cpu_to_le16(cid); + + return sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); +} + +/* send reset command by admin quueue temporary */ +int sssraid_send_reset_cmd(struct sssraid_dev *hdev, u8 type, u32 hdid) +{ + struct sssraid_admin_command admin_cmd; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.reset.opcode = SSSRAID_ADMIN_RESET; + admin_cmd.reset.hdid = cpu_to_le32(hdid); + admin_cmd.reset.type = type; + + return sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); +} + +static int sssraid_delete_queue(struct sssraid_dev *hdev, u8 op, u16 qidx) +{ + struct sssraid_admin_command admin_cmd; + int retval; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.delete_queue.opcode = op; + admin_cmd.delete_queue.qid = cpu_to_le16(qidx); + + retval = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); + + if (retval) + ioc_err(hdev, "Delete %s:[%d] failed\n", + (op == SSSRAID_ADMIN_DELETE_CQ) ? "cq" : "sq", qidx); + + return retval; +} + +static int sssraid_delete_cq(struct sssraid_dev *hdev, u16 qidx) +{ + return sssraid_delete_queue(hdev, SSSRAID_ADMIN_DELETE_CQ, qidx); +} + +void sssraid_adm_timeout(struct sssraid_dev *hdev, struct sssraid_cmd *cmd) +{ + /* command may be returned because controller reset */ + if (READ_ONCE(cmd->state) == SSSRAID_CMD_COMPLETE) + return; + + if (!sssraid_change_host_state(hdev, SSSRAID_RESETTING)) { + ioc_info(hdev, "can't change to reset state\n"); + return; + } + sssraid_soft_reset_handler(hdev); +} + +static int sssraid_create_io_qpair(struct sssraid_dev *hdev, u16 qidx) +{ + int retval; + + retval = sssraid_create_io_cq(hdev, qidx); + if (retval) + return retval; + + retval = sssraid_create_io_sq(hdev, qidx); + if (retval) + goto delete_cq; + + /* intr_info.msix_index substitute cq_vector */ + + /* io interrupt registry: + * not here, put above + */ + + sssraid_init_queue(hdev, qidx); + + return 0; + +delete_cq: + sssraid_delete_cq(hdev, qidx); + + return retval; +} + +static int sssraid_setup_io_qpair(struct sssraid_dev *hdev) +{ + u32 i, num_queues; + int retval = 0; + + num_queues = min(hdev->intr_info_count, hdev->init_done_queue_cnt - 1); + for (i = 1; i <= num_queues; i++) { + retval = sssraid_create_io_qpair(hdev, i); + if (retval) { + ioc_err(hdev, "Create queue[%d] failed\n", i); + break; + } + } + + ioc_info(hdev, "init_done_queue_cnt[%d], intr_info_count[%d] num_queues[%d]", + hdev->init_done_queue_cnt, /*hdev->online_queues,*/ + hdev->intr_info_count, num_queues); + + return retval >= 0 ? 0 : retval; +} + +static int sssraid_alloc_ioq_ptcmds(struct sssraid_dev *hdev) +{ + int i; + int ptnum = SSSRAID_NR_IOQ_PTCMDS; + + INIT_LIST_HEAD(&hdev->ioq_pt_list); + spin_lock_init(&hdev->ioq_pt_lock); + + hdev->ioq_ptcmds = kcalloc_node(ptnum, sizeof(struct sssraid_cmd), + GFP_KERNEL, hdev->numa_node); + + if (!hdev->ioq_ptcmds) { + ioc_err(hdev, "Alloc ioq_ptcmds failed\n"); + return -ENOMEM; + } + + for (i = 0; i < ptnum; i++) { + hdev->ioq_ptcmds[i].qid = i / SSSRAID_PTCMDS_PERQ + 1; + hdev->ioq_ptcmds[i].cid = i % SSSRAID_PTCMDS_PERQ + SSSRAID_IO_BLK_MQ_DEPTH; + list_add_tail(&(hdev->ioq_ptcmds[i].list), &hdev->ioq_pt_list); + } + + ioc_info(hdev, "Alloc ioq_ptcmds success, ptnum[%d]\n", ptnum); + + return 0; +} + +int sssraid_send_event_ack(struct sssraid_dev *hdev, u8 event, + u32 event_ctx, u16 cid) +{ + /* event,event_ctx no use at this time */ + struct sssraid_squeue *sqinfo = &hdev->sqinfo[0]; + struct sssraid_admin_command admin_cmd; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.common.opcode = SSSRAID_ADMIN_ASYNC_EVENT; + admin_cmd.common.command_id = cpu_to_le16(cid); + + sssraid_submit_cmd(sqinfo, &admin_cmd); + ioc_info(hdev, "send aen, cid[%d]\n", cid); + + return 0; +} + +static void sssraid_handle_aen_notice(struct sssraid_dev *hdev, u32 result) +{ + switch ((result & 0xff00) >> 8) { + case SSSRAID_AEN_DEV_CHANGED: + sssraid_scan_disk(hdev); + break; + case SSSRAID_AEN_FW_ACT_START: + ioc_info(hdev, "fw activation starting\n"); + break; + case SSSRAID_AEN_HOST_PROBING: + break; + default: + ioc_warn(hdev, "async event result %08x\n", result); + } +} + +static void sssraid_handle_aen_vs(struct sssraid_dev *hdev, u32 result, u32 result1) +{ + switch ((result & 0xff00) >> 8) { + case SSSRAID_AEN_TIMESYN: + sssraid_configure_timestamp(hdev); + break; + case SSSRAID_AEN_FW_ACT_FINISH: + ioc_info(hdev, "fw activation finish\n"); + if (sssraid_init_ctrl_info(hdev)) + ioc_err(hdev, "get ctrl info failed after fw act\n"); + break; + case SSSRAID_AEN_EVENT_MIN ... SSSRAID_AEN_EVENT_MAX: + ioc_info(hdev, "rcv card event[%d], param1[0x%x] param2[0x%x]\n", + (result & 0xff00) >> 8, result, result1); + break; + default: + ioc_warn(hdev, "async event result: 0x%x\n", result); + } +} + +static inline void sssraid_send_all_aen(struct sssraid_dev *hdev) +{ + u16 i; + + for (i = 0; i < hdev->ctrl_info->aerl; i++) + sssraid_send_event_ack(hdev, 0, 0, i + SSSRAID_AQ_BLK_MQ_DEPTH); +} + +static int sssraid_dev_list_init(struct sssraid_dev *hdev) +{ + u32 nd = le32_to_cpu(hdev->ctrl_info->nd); + + hdev->devices = kzalloc_node(nd * sizeof(struct sssraid_dev_info), + GFP_KERNEL, hdev->numa_node); + if (!hdev->devices) + return -ENOMEM; + + return 0; +} + +int sssraid_configure_timestamp(struct sssraid_dev *hdev) +{ + __le64 ts; + int retval; + + ts = cpu_to_le64(ktime_to_ms(ktime_get_real())); + retval = sssraid_set_features(hdev, SSSRAID_FEAT_TIMESTAMP, 0, &ts, sizeof(ts), NULL); + + if (retval) + ioc_err(hdev, "set timestamp failed: %d\n", retval); + return retval; +} + +int sssraid_init_ioc(struct sssraid_dev *hdev, u8 re_init) +{ + int retval = 0; + int i; + u32 nr_ioqs, bar_size; + + if (!re_init) { + hdev->cpu_count = num_online_cpus(); + + retval = sssraid_alloc_resources(hdev); + if (retval) { + ioc_err(hdev, "Failed to alloc resources:error %d\n", + retval); + goto out_nocleanup; + } + } + + /* reset need re-setup */ + retval = sssraid_setup_resources(hdev); + if (retval) { + ioc_err(hdev, "Failed to setup resources:error %d\n", + retval); + goto out_failed; + } + + if (!re_init) { + retval = sssraid_alloc_admin_cmds(hdev); + if (retval) { + ioc_err(hdev, "Failed to alloc admin cmds:error %d\n", + retval); + goto out_failed; + } + /* put here: + * alloc admin queue + */ + retval = sssraid_alloc_qpair(hdev, 0, SSSRAID_AQ_DEPTH); + if (retval) { + ioc_err(hdev, "Failed to alloc admin queue:error %d\n", + retval); + goto out_failed; + } + } + + retval = sssraid_setup_admin_qpair(hdev); + if (retval) { + ioc_err(hdev, "Failed to setup admin queue:error %d\n", + retval); + goto out_failed; + } + + /* 1. unregister all interrupt + * 2. admin interrupt registry + */ + retval = sssraid_setup_isr(hdev, 1); + if (retval) { + ioc_err(hdev, "Failed to setup ISR error %d\n", + retval); + goto out_failed; + } + + retval = sssraid_init_ctrl_info(hdev); + if (retval) { + ioc_err(hdev, "Failed to get ctrl info error %d\n", + retval); + goto out_failed; + } + + nr_ioqs = hdev->cpu_count; + retval = sssraid_set_queue_cnt(hdev, &nr_ioqs); + if (retval) { + ioc_err(hdev, "Failed to set queue cnt error %d\n", + retval); + goto out_failed; + } + + hdev->before_affinity_msix_cnt = nr_ioqs + 1; + + /* 1. unregister all interrupt + * 2. admin interrupt re-registry + * 3. io interrupt registry + */ + retval = sssraid_setup_isr(hdev, 0); + if (retval) { + ioc_err(hdev, "Failed to re-setup ISR, error %d\n", + retval); + goto out_failed; + } + + /* remap */ + bar_size = SSSRAID_REG_DBS + ((nr_ioqs + 1) * 8 * hdev->db_stride); + retval = sssraid_remap_bar(hdev, bar_size); + if (retval) { + ioc_err(hdev, "Failed to re-map bar, error %d\n", + retval); + goto out_failed; + } + hdev->sqinfo[0].q_db = hdev->dbs; + + /* num_vecs no sense, abandon */ + + if (!re_init) { + for (i = hdev->init_done_queue_cnt; i <= hdev->intr_info_count; i++) { + retval = sssraid_alloc_qpair(hdev, i, hdev->ioq_depth); + if (retval) { + ioc_err(hdev, "Failed to alloc io queue:error %d\n", + retval); + goto out_failed; + } + } + ioc_info(hdev, "intr_info_count: %d, init_done_queue_cnt: %d, ioq_depth: %d\n", + hdev->intr_info_count, hdev->init_done_queue_cnt, hdev->ioq_depth); + } + + retval = sssraid_setup_io_qpair(hdev); + if (retval) { + ioc_err(hdev, "Failed to setup io qpair, error %d\n", + retval); + goto out_failed; + } + + if (!re_init) { + retval = sssraid_alloc_ioq_ptcmds(hdev); + if (retval) { + ioc_err(hdev, "Failed to alloc ioq ptcmds, error %d\n", + retval); + goto out_failed; + } + } + + sssraid_send_all_aen(hdev); + + if (!re_init) { + retval = sssraid_dev_list_init(hdev); + if (retval) { + ioc_err(hdev, "Failed to init device list, error %d\n", + retval); + goto out_failed; + } + + retval = sssraid_configure_timestamp(hdev); + if (retval) { + ioc_err(hdev, "Failed to configure timestamp, error %d\n", + retval); + goto out_failed; + } + } + + return retval; + +out_failed: + sssraid_cleanup_ioc(hdev, re_init); +out_nocleanup: + return retval; +} + +void sssraid_cleanup_resources(struct sssraid_dev *hdev) +{ + struct pci_dev *pdev = hdev->pdev; + + sssraid_cleanup_isr(hdev); + + if (hdev->bar) { + iounmap(hdev->bar); + hdev->bar = NULL; + } + + if (pci_is_enabled(pdev)) { + pci_disable_pcie_error_reporting(pdev); + pci_release_mem_regions(pdev); + pci_disable_device(pdev); + } +} + +static void sssraid_free_ioq_ptcmds(struct sssraid_dev *hdev) +{ + kfree(hdev->ioq_ptcmds); + hdev->ioq_ptcmds = NULL; + + INIT_LIST_HEAD(&hdev->ioq_pt_list); +} + +static void sssraid_delete_io_queues(struct sssraid_dev *hdev) +{ + u16 queues = hdev->init_done_queue_cnt - 1; + u8 opcode = SSSRAID_ADMIN_DELETE_SQ; + u16 i, pass; + + if (!pci_device_is_present(hdev->pdev)) { + ioc_err(hdev, "pci_device is not present, skip disable io queues\n"); + return; + } + + if (hdev->init_done_queue_cnt < 2) { + ioc_err(hdev, "err, io queue has been delete\n"); + return; + } + + for (pass = 0; pass < 2; pass++) { + for (i = queues; i > 0; i--) + if (sssraid_delete_queue(hdev, opcode, i)) + break; + + opcode = SSSRAID_ADMIN_DELETE_CQ; + } +} + +void sssraid_complete_aen(struct sssraid_dev *hdev, struct sssraid_completion *cqe) +{ + u32 result = le32_to_cpu(cqe->result); + + ioc_info(hdev, "rcv aen, cid[%d], status[0x%x], result[0x%x]\n", + le16_to_cpu(cqe->cmd_id), le16_to_cpu(cqe->status) >> 1, result); + + /* + * The response to event moved from this func. + * sssraid_send_aen changed to name sssraid_send_event_ack + */ + + if ((le16_to_cpu(cqe->status) >> 1) != SSSRAID_SC_SUCCESS) + return; + switch (result & 0x7) { + case SSSRAID_AEN_NOTICE: + sssraid_handle_aen_notice(hdev, result); + break; + case SSSRAID_AEN_VS: + sssraid_handle_aen_vs(hdev, result, le32_to_cpu(cqe->result1)); + break; + default: + ioc_warn(hdev, "Unsupported async event type: %u\n", + result & 0x7); + break; + } +} + +void sssraid_free_iod_res(struct sssraid_dev *hdev, struct sssraid_iod *iod) +{ + const int last_prp = hdev->page_size / sizeof(__le64) - 1; + dma_addr_t dma_addr, next_dma_addr; + struct sssraid_sgl_desc *sg_list; + __le64 *prp_list; + void *addr; + int i; + + dma_addr = iod->first_dma; + if (iod->npages == 0) + dma_pool_free(iod->sqinfo->prp_small_pool, sssraid_iod_list(iod)[0], dma_addr); + + for (i = 0; i < iod->npages; i++) { + addr = sssraid_iod_list(iod)[i]; + + if (iod->use_sgl) { + sg_list = addr; + next_dma_addr = + le64_to_cpu((sg_list[SGES_PER_PAGE - 1]).addr); + } else { + prp_list = addr; + next_dma_addr = le64_to_cpu(prp_list[last_prp]); + } + + dma_pool_free(hdev->prp_page_pool, addr, dma_addr); + dma_addr = next_dma_addr; + } + + iod->sense = NULL; + iod->npages = -1; +} + +static void sssraid_complete_ioq_sync_cmnd(struct sssraid_dev *hdev, u16 qidx, + struct sssraid_completion *cqe) +{ + struct sssraid_cmd *ptcmd; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + + ptcmd = hdev->ioq_ptcmds + (sqinfo->qidx - 1) * SSSRAID_PTCMDS_PERQ + + le16_to_cpu(cqe->cmd_id) - SSSRAID_IO_BLK_MQ_DEPTH; + + ptcmd->status = le16_to_cpu(cqe->status) >> 1; + ptcmd->result0 = le32_to_cpu(cqe->result); + ptcmd->result1 = le32_to_cpu(cqe->result1); + + complete(&ptcmd->cmd_done); +} + +static void sssraid_complete_ioq_cmnd(struct sssraid_dev *hdev, u16 qidx, + struct sssraid_completion *cqe) +{ + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + + struct blk_mq_tags *tags; + struct scsi_cmnd *scmd; + struct sssraid_iod *iod; + struct request *req; + unsigned long elapsed; + + tags = hdev->shost->tag_set.tags[sqinfo->qidx - 1]; + + req = blk_mq_tag_to_rq(tags, le16_to_cpu(cqe->cmd_id)); + if (unlikely(!req || !blk_mq_request_started(req))) { + ioc_warn(hdev, "Invalid id %d completed on queue %d\n", + le16_to_cpu(cqe->cmd_id), sqinfo->qidx); + return; + } + + scmd = blk_mq_rq_to_pdu(req); + iod = scsi_cmd_priv(scmd); + + elapsed = jiffies - scmd->jiffies_at_alloc; + dbgprint(hdev, "cid[%d] qidx[%d] finish IO cost %3ld.%3ld seconds\n", + le16_to_cpu(cqe->cmd_id), sqinfo->qidx, elapsed / HZ, elapsed % HZ); + + if (cmpxchg(&iod->state, SSSRAID_CMD_IN_FLIGHT, SSSRAID_CMD_COMPLETE) != + SSSRAID_CMD_IN_FLIGHT) { + ioc_warn(hdev, "cid[%d] qidx[%d] enters abnormal handler, cost %3ld.%3ld seconds\n", + le16_to_cpu(cqe->cmd_id), sqinfo->qidx, elapsed / HZ, elapsed % HZ); + WRITE_ONCE(iod->state, SSSRAID_CMD_TMO_COMPLETE); + + if (iod->nsge) { + iod->nsge = 0; + scsi_dma_unmap(scmd); + } + sssraid_free_iod_res(hdev, iod); + + return; + } + + sssraid_map_status(iod, scmd, cqe); + if (iod->nsge) { + iod->nsge = 0; + scsi_dma_unmap(scmd); + } + sssraid_free_iod_res(hdev, iod); + scmd->scsi_done(scmd); +} + +static void sssraid_process_admin_cq(struct sssraid_dev *hdev, + struct sssraid_squeue *sqinfo, + struct sssraid_completion *cqe) +{ + struct sssraid_fwevt *fwevt = NULL; + u16 cid = le16_to_cpu(cqe->cmd_id), sz; + + if (likely(cid < SSSRAID_AQ_BLK_MQ_DEPTH)) + sssraid_complete_adminq_cmnd(hdev, sqinfo->qidx, cqe); + else { + sz = sizeof(*cqe); + fwevt = sssraid_alloc_fwevt(sz); + if (!fwevt) { + ioc_info(hdev, "%s :failure at %s:%d/%s()!\n", + __func__, __FILE__, __LINE__, __func__); + return; + } + + memcpy(fwevt->event_data, cqe, sz); + fwevt->hdev = hdev; + fwevt->event_id = 0; /* evt_type:0 */ + fwevt->send_ack = 1; /* ack_req:1 */ + fwevt->process_evt = 1; /* process_evt_bh:1 */ + fwevt->evt_ctx = 0; /* 0 */ + sssraid_fwevt_add_to_list(hdev, fwevt); + } +} + +static void sssraid_process_io_cq(struct sssraid_dev *hdev, + struct sssraid_squeue *sqinfo, + struct sssraid_completion *cqe) +{ + u16 cid = le16_to_cpu(cqe->cmd_id); + + if (likely(cid < SSSRAID_IO_BLK_MQ_DEPTH)) + sssraid_complete_ioq_cmnd(hdev, sqinfo->qidx, cqe); + else /* io sync handle */ + sssraid_complete_ioq_sync_cmnd(hdev, sqinfo->qidx, cqe); +} + +static inline void sssraid_handle_cqe(struct sssraid_dev *hdev, u16 mdix, u16 didx) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[mdix]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[mdix]; + + struct sssraid_completion *cqe = &cqinfo->cqes[didx]; + u16 cid = le16_to_cpu(cqe->cmd_id); + + if (unlikely(cid >= sqinfo->q_depth)) { + ioc_err(hdev, "Invalid command id[%d] completed on queue %d\n", + cid, cqe->sq_id); + return; + } + + dbgprint(hdev, "cid[%d] mdix[%d], result[0x%x], sq_id[%d], status[0x%x]\n", + cid, sqinfo->qidx, le32_to_cpu(cqe->result), + le16_to_cpu(cqe->sq_id), le16_to_cpu(cqe->status)); + + if (!mdix) /* admin */ + sssraid_process_admin_cq(hdev, sqinfo, cqe); + else /* io */ + sssraid_process_io_cq(hdev, sqinfo, cqe); +} + +void sssraid_complete_cqes(struct sssraid_dev *hdev, u16 midx, u16 start, u16 end) +{ + struct sssraid_squeue *sqinfo = &hdev->sqinfo[midx]; + + while (start != end) { + sssraid_handle_cqe(hdev, midx, start); + if (++start == sqinfo->q_depth) + start = 0; + } +} + +static void sssraid_disable_admin_queue(struct sssraid_dev *hdev, bool shutdown) +{ + struct sssraid_cqueue *adm_cqinfo = &hdev->cqinfo[0]; + u16 start, end; + + if (pci_device_is_present(hdev->pdev)) { + if (shutdown) + sssraid_shutdown_ctrl(hdev); + else + sssraid_disable_ctrl(hdev); + } + + if (hdev->init_done_queue_cnt == 0) { + ioc_err(hdev, "err, admin queue has been delete\n"); + return; + } + + spin_lock_irq(&adm_cqinfo->cq_lock); + sssraid_process_cq(hdev, 0, &start, &end, -1); + spin_unlock_irq(&adm_cqinfo->cq_lock); + sssraid_complete_cqes(hdev, 0, start, end); +} + +static void sssraid_free_all_queues(struct sssraid_dev *hdev) +{ + int i; + struct sssraid_cqueue *cqinfo; + struct sssraid_squeue *sqinfo; + + for (i = 0; i < hdev->init_done_queue_cnt; i++) { + cqinfo = &hdev->cqinfo[i]; + sqinfo = &hdev->sqinfo[i]; + dma_free_coherent(&hdev->pdev->dev, CQ_SIZE(sqinfo->q_depth), + (void *)cqinfo->cqes, cqinfo->cq_dma_addr); + dma_free_coherent(&hdev->pdev->dev, SQ_SIZE(sqinfo->qidx, sqinfo->q_depth), + sqinfo->sq_cmds, sqinfo->sq_dma_addr); + dma_free_coherent(&hdev->pdev->dev, SENSE_SIZE(sqinfo->q_depth), + sqinfo->sense, sqinfo->sense_dma_addr); + } + + hdev->init_done_queue_cnt = 0; +} + +static void sssraid_free_admin_cmds(struct sssraid_dev *hdev) +{ + kfree(hdev->adm_cmds); + hdev->adm_cmds = NULL; + INIT_LIST_HEAD(&hdev->adm_cmd_list); +} + +static void sssraid_free_resources(struct sssraid_dev *hdev) +{ + sssraid_free_admin_cmds(hdev); + kfree(hdev->sqinfo); + kfree(hdev->cqinfo); + sssraid_destroy_dma_pools(hdev); + kfree(hdev->ctrl_info); +} + +void sssraid_cleanup_ioc(struct sssraid_dev *hdev, u8 re_init) +{ + if (!re_init) + sssraid_free_ioq_ptcmds(hdev); + + sssraid_delete_io_queues(hdev); + sssraid_disable_admin_queue(hdev, !re_init); + + if (!re_init) + sssraid_free_all_queues(hdev); + + sssraid_ioc_disable_intr(hdev); + sssraid_cleanup_resources(hdev); + + if (!re_init) + sssraid_free_resources(hdev); + +} + +int sssraid_soft_reset_handler(struct sssraid_dev *hdev) +{ + int retval = 0; + + if (hdev->state != SSSRAID_RESETTING) { + ioc_err(hdev, "err, host is not reset state\n"); + return retval; + } + + ioc_info(hdev, "enter host reset\n"); + + sssraid_ioc_disable_intr(hdev); + sssraid_cleanup_fwevt_list(hdev); + + /* realize above here: + * sssraid_dev_disable -> sssraid_back_all_io + */ + sssraid_back_all_io(hdev); + /* + * realize sssraid_dev_disable, + * i.e. sssraid_cleanup_ioc(1) + */ + if (hdev->ctrl_config & SSSRAID_CC_ENABLE) { + ioc_info(hdev, "start dev_disable\n"); + sssraid_cleanup_ioc(hdev, 1); + } + + retval = sssraid_init_ioc(hdev, 1); + if (retval) { + ioc_err(hdev, "init ioc fail.\n"); + return retval; + } + + sssraid_change_host_state(hdev, SSSRAID_LIVE); + + return retval; +} diff --git a/drivers/scsi/sssraid/sssraid_os.c b/drivers/scsi/sssraid/sssraid_os.c new file mode 100644 index 000000000000..16d2b94d78f3 --- /dev/null +++ b/drivers/scsi/sssraid/sssraid_os.c @@ -0,0 +1,2337 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2022 3SNIC Information Technology, Ltd */ + +/* 3SNIC RAID SSSXXX Series Linux Driver */ + +#include <linux/version.h> +#include <linux/pci.h> +#include <linux/aer.h> +#include <linux/module.h> +#include <linux/ioport.h> +#include <linux/device.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/cdev.h> +#include <linux/sysfs.h> +#include <linux/gfp.h> +#include <linux/types.h> +#include <linux/ratelimit.h> +#include <linux/debugfs.h> +#include <linux/blkdev.h> +#include <linux/bsg-lib.h> +#include <linux/sort.h> + +#include <scsi/scsi.h> +#include <scsi/scsi_cmnd.h> +#include <scsi/scsi_device.h> +#include <scsi/scsi_host.h> +#include <scsi/scsi_transport.h> +#include <scsi/scsi_dbg.h> + +#include <linux/unaligned/be_byteshift.h> +#include <linux/unaligned/le_byteshift.h> +#include <linux/once.h> +#include <linux/sched/signal.h> +#include <linux/io-64-nonatomic-lo-hi.h> + +#include "sssraid.h" +#include "sssraid_debug.h" + +u32 admin_tmout = 60; +module_param(admin_tmout, uint, 0644); +MODULE_PARM_DESC(admin_tmout, "admin commands timeout (seconds)"); + +static u32 scmd_tmout_rawdisk = 180; +module_param(scmd_tmout_rawdisk, uint, 0644); +MODULE_PARM_DESC(scmd_tmout_rawdisk, "scsi commands timeout for rawdisk(seconds)"); + +static u32 scmd_tmout_vd = 180; +module_param(scmd_tmout_vd, uint, 0644); +MODULE_PARM_DESC(scmd_tmout_vd, "scsi commands timeout for vd(seconds)"); + +static int ioq_depth_set(const char *val, const struct kernel_param *kp); +static const struct kernel_param_ops ioq_depth_ops = { + .set = ioq_depth_set, + .get = param_get_uint, +}; + +u32 io_queue_depth = 1024; +module_param_cb(io_queue_depth, &ioq_depth_ops, &io_queue_depth, 0644); +MODULE_PARM_DESC(io_queue_depth, "set io queue depth, should >= 2"); + +static int logging_level_set(const char *val, const struct kernel_param *kp) +{ + u8 n = 0; + int ret; + + ret = kstrtou8(val, 10, &n); + if (ret != 0) + return -EINVAL; + + return param_set_byte(val, kp); +} + +static const struct kernel_param_ops logging_level_ops = { + .set = logging_level_set, + .get = param_get_byte, +}; + +static unsigned char logging_level; +module_param_cb(logging_level, &logging_level_ops, &logging_level, 0644); +MODULE_PARM_DESC(logging_level, "set log level, default zero for switch off"); + +static int small_pool_num_set(const char *val, const struct kernel_param *kp) +{ + u8 n = 0; + int ret; + + ret = kstrtou8(val, 10, &n); + if (ret != 0) + return -EINVAL; + if (n > MAX_SMALL_POOL_NUM) + n = MAX_SMALL_POOL_NUM; + if (n < 1) + n = 1; + *((u8 *)kp->arg) = n; + + return 0; +} + +static const struct kernel_param_ops small_pool_num_ops = { + .set = small_pool_num_set, + .get = param_get_byte, +}; + +/* Small pools are used to save PRP for small IOs.It was + * found that the spinlock of a single pool conflicts a + * lot with multiple CPUs.So multiple pools are introduced + * to reduce the conflictions. + */ +unsigned char small_pool_num = 4; +module_param_cb(small_pool_num, &small_pool_num_ops, &small_pool_num, 0644); +MODULE_PARM_DESC(small_pool_num, "set prp small pool num, default 4, MAX 16"); + +//static struct class *sssraid_class; + +enum FW_STAT_CODE { + FW_STAT_OK = 0, + FW_STAT_NEED_CHECK, + FW_STAT_ERROR, + FW_STAT_EP_PCIE_ERROR, + FW_STAT_NAC_DMA_ERROR, + FW_STAT_ABORTED, + FW_STAT_NEED_RETRY +}; + +static const char * const raid_levels[] = {"0", "1", "5", "6", "10", "50", "60", "NA"}; + +static const char * const raid_states[] = { + "NA", "NORMAL", "FAULT", "DEGRADE", "NOT_FORMATTED", "FORMATTING", "SANITIZING", + "INITIALIZING", "INITIALIZE_FAIL", "DELETING", "DELETE_FAIL", "WRITE_PROTECT" +}; + +static int ioq_depth_set(const char *val, const struct kernel_param *kp) +{ + int n = 0; + int ret; + + ret = kstrtoint(val, 10, &n); + if (ret != 0 || n < 2) + return -EINVAL; + + return param_set_int(val, kp); +} + +/* + * common + */ +static struct class *sssraid_class; + +struct sssraid_fwevt *sssraid_alloc_fwevt(int len) +{ + struct sssraid_fwevt *fwevt; + + fwevt = kzalloc(sizeof(*fwevt) + len, GFP_ATOMIC); + if (!fwevt) + return NULL; + + kref_init(&fwevt->ref_count); + return fwevt; + +} + +static void sssraid_fwevt_free(struct kref *r) +{ + kfree(container_of(r, struct sssraid_fwevt, ref_count)); +} + +static void sssraid_fwevt_get(struct sssraid_fwevt *fwevt) +{ + kref_get(&fwevt->ref_count); +} + +static void sssraid_fwevt_put(struct sssraid_fwevt *fwevt) +{ + kref_put(&fwevt->ref_count, sssraid_fwevt_free); +} + +static void sssraid_fwevt_del_from_list(struct sssraid_dev *hdev, + struct sssraid_fwevt *fwevt) +{ + unsigned long flags; + + spin_lock_irqsave(&hdev->fwevt_lock, flags); + if (!list_empty(&fwevt->list)) { + list_del_init(&fwevt->list); + /* + * Put fwevt reference count after + * removing it from fwevt_list + */ + sssraid_fwevt_put(fwevt); + } + spin_unlock_irqrestore(&hdev->fwevt_lock, flags); +} + +static void sssraid_fwevt_bh(struct sssraid_dev *hdev, + struct sssraid_fwevt *fwevt) +{ + struct sssraid_completion *cqe; + + hdev->current_event = fwevt; + sssraid_fwevt_del_from_list(hdev, fwevt); + + cqe = (struct sssraid_completion *)fwevt->event_data; + + if (!fwevt->process_evt) + goto evt_ack; + + sssraid_complete_aen(hdev, cqe); + +evt_ack: + /* event response put here: event has been handled. */ + sssraid_send_event_ack(hdev, fwevt->event_id, + fwevt->evt_ctx, le16_to_cpu(cqe->cmd_id)); + sssraid_fwevt_put(fwevt); + hdev->current_event = NULL; +} + +static void sssraid_fwevt_worker(struct work_struct *work) +{ + struct sssraid_fwevt *fwevt = container_of(work, struct sssraid_fwevt, + work); + sssraid_fwevt_bh(fwevt->hdev, fwevt); + /* + * Put fwevt reference count after + * dequeuing it from worker queue + */ + sssraid_fwevt_put(fwevt); +} + +void sssraid_fwevt_add_to_list(struct sssraid_dev *hdev, + struct sssraid_fwevt *fwevt) +{ + unsigned long flags; + + if (!hdev->fwevt_worker_thread) + return; + + spin_lock_irqsave(&hdev->fwevt_lock, flags); + /* get fwevt reference count while adding it to fwevt_list */ + sssraid_fwevt_get(fwevt); + INIT_LIST_HEAD(&fwevt->list); + list_add_tail(&fwevt->list, &hdev->fwevt_list); + INIT_WORK(&fwevt->work, sssraid_fwevt_worker); + /* get fwevt reference count while enqueueing it to worker queue */ + sssraid_fwevt_get(fwevt); + queue_work(hdev->fwevt_worker_thread, &fwevt->work); + spin_unlock_irqrestore(&hdev->fwevt_lock, flags); +} + +static struct sssraid_fwevt *sssraid_dequeue_fwevt( + struct sssraid_dev *hdev) +{ + unsigned long flags; + struct sssraid_fwevt *fwevt = NULL; + + spin_lock_irqsave(&hdev->fwevt_lock, flags); + if (!list_empty(&hdev->fwevt_list)) { + fwevt = list_first_entry(&hdev->fwevt_list, + struct sssraid_fwevt, list); + list_del_init(&fwevt->list); + /* + * Put fwevt reference count after + * removing it from fwevt_list + */ + sssraid_fwevt_put(fwevt); + } + spin_unlock_irqrestore(&hdev->fwevt_lock, flags); + + return fwevt; +} + +void sssraid_cleanup_fwevt_list(struct sssraid_dev *hdev) +{ + struct sssraid_fwevt *fwevt = NULL; + + if ((list_empty(&hdev->fwevt_list) && !hdev->current_event) || + !hdev->fwevt_worker_thread) + return; + + while ((fwevt = sssraid_dequeue_fwevt(hdev)) || + (fwevt = hdev->current_event)) { + /* + * Wait on the fwevt to complete. If this returns 1, then + * the event was never executed, and we need a put for the + * reference the work had on the fwevt. + * + * If it did execute, we wait for it to finish, and the put will + * happen from sssraid_process_fwevt() + */ + if (cancel_work_sync(&fwevt->work)) { + /* + * Put fwevt reference count after + * dequeuing it from worker queue + */ + sssraid_fwevt_put(fwevt); + /* + * Put fwevt reference count to neutralize + * kref_init increment + */ + sssraid_fwevt_put(fwevt); + } + } +} + +/* + * common 1 + */ +static int sssraid_npages_prp(struct sssraid_dev *hdev) +{ + u32 size = 1U << ((hdev->ctrl_info->mdts) * 1U) << 12; + u32 nprps = DIV_ROUND_UP(size + hdev->page_size, hdev->page_size); + + return DIV_ROUND_UP(PRP_ENTRY_SIZE * nprps, hdev->page_size - PRP_ENTRY_SIZE); +} + +static int sssraid_npages_sgl(struct sssraid_dev *hdev) +{ + u32 nsge = le16_to_cpu(hdev->ctrl_info->max_num_sge); + + return DIV_ROUND_UP(nsge * sizeof(struct sssraid_sgl_desc), hdev->page_size); +} + +static u32 sssraid_cmd_size(struct sssraid_dev *hdev) +{ + u32 alloc_size = sizeof(__le64 *) * max(sssraid_npages_prp(hdev), sssraid_npages_sgl(hdev)); + + ioc_info(hdev, "iod size: %lu, alloc_size: %u\n", + sizeof(struct sssraid_iod), alloc_size); + + return sizeof(struct sssraid_iod) + alloc_size; +} + +static int sssraid_setup_prps(struct sssraid_dev *hdev, struct sssraid_iod *iod) +{ + struct scatterlist *sg = iod->sg; + u64 dma_addr = sg_dma_address(sg); + int dma_len = sg_dma_len(sg); + __le64 *prp_list, *old_prp_list; + int page_size = hdev->page_size; + int offset = dma_addr & (page_size - 1); + void **list = sssraid_iod_list(iod); + int length = iod->length; + struct dma_pool *pool; + dma_addr_t prp_dma; + int nprps, i; + + length -= (page_size - offset); + if (length <= 0) { + iod->first_dma = 0; + return 0; + } + + dma_len -= (page_size - offset); + if (dma_len) { + dma_addr += (page_size - offset); + } else { + sg = sg_next(sg); + dma_addr = sg_dma_address(sg); + dma_len = sg_dma_len(sg); + } + + if (length <= page_size) { + iod->first_dma = dma_addr; + return 0; + } + + nprps = DIV_ROUND_UP(length, page_size); + if (nprps <= (SMALL_POOL_SIZE / PRP_ENTRY_SIZE)) { + pool = iod->sqinfo->prp_small_pool; + iod->npages = 0; + } else { + pool = hdev->prp_page_pool; + iod->npages = 1; + } + + prp_list = dma_pool_alloc(pool, GFP_ATOMIC, &prp_dma); + if (!prp_list) { + dev_err_ratelimited(&hdev->pdev->dev, "Allocate first prp_list memory failed\n"); + iod->first_dma = dma_addr; + iod->npages = -1; + return -ENOMEM; + } + list[0] = prp_list; + iod->first_dma = prp_dma; + i = 0; + for (;;) { + if (i == page_size / PRP_ENTRY_SIZE) { + old_prp_list = prp_list; + + prp_list = dma_pool_alloc(pool, GFP_ATOMIC, &prp_dma); + if (!prp_list) { + dev_err_ratelimited(&hdev->pdev->dev, "Allocate %dth prp_list memory failed\n", + iod->npages + 1); + return -ENOMEM; + } + list[iod->npages++] = prp_list; + prp_list[0] = old_prp_list[i - 1]; + old_prp_list[i - 1] = cpu_to_le64(prp_dma); + i = 1; + } + prp_list[i++] = cpu_to_le64(dma_addr); + dma_len -= page_size; + dma_addr += page_size; + length -= page_size; + if (length <= 0) + break; + if (dma_len > 0) + continue; + if (unlikely(dma_len < 0)) + goto bad_sgl; + sg = sg_next(sg); + dma_addr = sg_dma_address(sg); + dma_len = sg_dma_len(sg); + } + + return 0; + +bad_sgl: + ioc_err(hdev, "Setup prps, invalid SGL for payload: %d nents: %d\n", + iod->length, iod->nsge); + return -EIO; +} + +static inline bool sssraid_is_rw_scmd(struct scsi_cmnd *scmd) +{ + switch (scmd->cmnd[0]) { + case READ_6: + case READ_10: + case READ_12: + case READ_16: + case WRITE_6: + case WRITE_10: + case WRITE_12: + case WRITE_16: + return true; + default: + return false; + } +} + +static bool sssraid_is_prp(struct sssraid_dev *hdev, struct scsi_cmnd *scmd, u32 nsge) +{ + struct scatterlist *sg = scsi_sglist(scmd); + u32 page_mask = hdev->page_size - 1; + bool is_prp = true; + int i = 0; + + scsi_for_each_sg(scmd, sg, nsge, i) { + if (i != 0 && i != nsge - 1) { + if ((sg_dma_len(sg) & page_mask) || + (sg_dma_address(sg) & page_mask)) { + is_prp = false; + break; + } + } + + if (nsge > 1 && i == 0) { + if ((sg_dma_address(sg) + sg_dma_len(sg)) & page_mask) { + is_prp = false; + break; + } + } + + if (nsge > 1 && i == (nsge - 1)) { + if (sg_dma_address(sg) & page_mask) { + is_prp = false; + break; + } + } + } + + return is_prp; +} + +static void sssraid_sgl_set_data(struct sssraid_sgl_desc *sge, struct scatterlist *sg) +{ + sge->addr = cpu_to_le64(sg_dma_address(sg)); + sge->length = cpu_to_le32(sg_dma_len(sg)); + sge->type = SSSRAID_SGL_FMT_DATA_DESC << 4; +} + +static void sssraid_sgl_set_seg(struct sssraid_sgl_desc *sge, dma_addr_t dma_addr, int entries) +{ + sge->addr = cpu_to_le64(dma_addr); + if (entries <= SGES_PER_PAGE) { + sge->length = cpu_to_le32(entries * sizeof(*sge)); + sge->type = SSSRAID_SGL_FMT_LAST_SEG_DESC << 4; + } else { + sge->length = cpu_to_le32(PAGE_SIZE); + sge->type = SSSRAID_SGL_FMT_SEG_DESC << 4; + } +} + + +static int sssraid_setup_ioq_cmd_sgl(struct sssraid_dev *hdev, + struct scsi_cmnd *scmd, struct sssraid_ioq_command *ioq_cmd, + struct sssraid_iod *iod) +{ + struct sssraid_sgl_desc *sg_list, *link, *old_sg_list; + struct scatterlist *sg = scsi_sglist(scmd); + void **list = sssraid_iod_list(iod); + struct dma_pool *pool; + int nsge = iod->nsge; + dma_addr_t sgl_dma; + int i = 0; + + ioq_cmd->common.flags |= SSSRAID_CMD_FLAG_SGL_METABUF; + + if (nsge == 1) { + sssraid_sgl_set_data(&ioq_cmd->common.dptr.sgl, sg); + return 0; + } + + if (nsge <= (SMALL_POOL_SIZE / sizeof(struct sssraid_sgl_desc))) { + pool = iod->sqinfo->prp_small_pool; + iod->npages = 0; + } else { + pool = hdev->prp_page_pool; + iod->npages = 1; + } + + sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &sgl_dma); + if (!sg_list) { + dev_err_ratelimited(&hdev->pdev->dev, "Allocate first sgl_list failed\n"); + iod->npages = -1; + return -ENOMEM; + } + + list[0] = sg_list; + iod->first_dma = sgl_dma; + sssraid_sgl_set_seg(&ioq_cmd->common.dptr.sgl, sgl_dma, nsge); + do { + if (i == SGES_PER_PAGE) { + old_sg_list = sg_list; + link = &old_sg_list[SGES_PER_PAGE - 1]; + + sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &sgl_dma); + if (!sg_list) { + dev_err_ratelimited(&hdev->pdev->dev, "Allocate %dth sgl_list failed\n", + iod->npages + 1); + return -ENOMEM; + } + list[iod->npages++] = sg_list; + + i = 0; + memcpy(&sg_list[i++], link, sizeof(*link)); + sssraid_sgl_set_seg(link, sgl_dma, nsge); + } + + sssraid_sgl_set_data(&sg_list[i++], sg); + sg = sg_next(sg); + } while (--nsge > 0); + + return 0; +} + +static void sssraid_shost_init(struct sssraid_dev *hdev) +{ + struct pci_dev *pdev = hdev->pdev; + u8 domain, bus; + u32 dev_func; + + domain = pci_domain_nr(pdev->bus); + bus = pdev->bus->number; + dev_func = pdev->devfn; + + hdev->shost->nr_hw_queues = hdev->init_done_queue_cnt - 1; + hdev->shost->can_queue = (hdev->ioq_depth - SSSRAID_PTCMDS_PERQ); + + hdev->shost->sg_tablesize = le16_to_cpu(hdev->ctrl_info->max_num_sge); + /* 512B per sector */ + hdev->shost->max_sectors = (1U << ((hdev->ctrl_info->mdts) * 1U) << 12) / 512; + hdev->shost->cmd_per_lun = MAX_CMD_PER_DEV; + hdev->shost->max_channel = le16_to_cpu(hdev->ctrl_info->max_channel) - 1; + hdev->shost->max_id = le32_to_cpu(hdev->ctrl_info->max_tgt_id); + hdev->shost->max_lun = le16_to_cpu(hdev->ctrl_info->max_lun); + + hdev->shost->this_id = -1; + hdev->shost->unique_id = (domain << 16) | (bus << 8) | dev_func; + hdev->shost->max_cmd_len = MAX_CDB_LEN; + hdev->shost->hostt->cmd_size = sssraid_cmd_size(hdev); +} + +static inline void sssraid_get_tag_from_scmd(struct scsi_cmnd *scmd, u16 *qidx, u16 *cid) +{ + u32 tag = blk_mq_unique_tag(scmd->request); + + *qidx = blk_mq_unique_tag_to_hwq(tag) + 1; + *cid = blk_mq_unique_tag_to_tag(tag); +} + +static inline uint32_t get_unaligned_be24(const uint8_t *const p) +{ + return get_unaligned_be32(p - 1) & 0xffffffU; +} + +static int sssraid_setup_rw_cmd(struct sssraid_dev *hdev, + struct sssraid_rw_command *rw, + struct scsi_cmnd *scmd) +{ + u32 start_lba_lo, start_lba_hi; + u32 datalength = 0; + u16 control = 0; + + start_lba_lo = 0; + start_lba_hi = 0; + + if (scmd->sc_data_direction == DMA_TO_DEVICE) { + rw->opcode = SSSRAID_CMD_WRITE; + } else if (scmd->sc_data_direction == DMA_FROM_DEVICE) { + rw->opcode = SSSRAID_CMD_READ; + } else { + ioc_err(hdev, "Invalid RW_IO for unsupported data direction: %d\n", + scmd->sc_data_direction); + WARN_ON(1); + return -EINVAL; + } + + /* 6-byte READ(0x08) or WRITE(0x0A) cdb */ + if (scmd->cmd_len == 6) { + datalength = (u32)(scmd->cmnd[4] == 0 ? + IO_6_DEFAULT_TX_LEN : scmd->cmnd[4]); + start_lba_lo = (u32)get_unaligned_be24(&scmd->cmnd[1]); + + start_lba_lo &= 0x1FFFFF; + } + + /* 10-byte READ(0x28) or WRITE(0x2A) cdb */ + else if (scmd->cmd_len == 10) { + datalength = (u32)get_unaligned_be16(&scmd->cmnd[7]); + start_lba_lo = get_unaligned_be32(&scmd->cmnd[2]); + + if (scmd->cmnd[1] & FUA_MASK) + control |= SSSRAID_RW_FUA; + } + + /* 12-byte READ(0xA8) or WRITE(0xAA) cdb */ + else if (scmd->cmd_len == 12) { + datalength = get_unaligned_be32(&scmd->cmnd[6]); + start_lba_lo = get_unaligned_be32(&scmd->cmnd[2]); + + if (scmd->cmnd[1] & FUA_MASK) + control |= SSSRAID_RW_FUA; + } + /* 16-byte READ(0x88) or WRITE(0x8A) cdb */ + else if (scmd->cmd_len == 16) { + datalength = get_unaligned_be32(&scmd->cmnd[10]); + start_lba_lo = get_unaligned_be32(&scmd->cmnd[6]); + start_lba_hi = get_unaligned_be32(&scmd->cmnd[2]); + + if (scmd->cmnd[1] & FUA_MASK) + control |= SSSRAID_RW_FUA; + } + + if (unlikely(datalength > U16_MAX || datalength == 0)) { + ioc_err(hdev, "Invalid IO for illegal transfer data length: %u\n", datalength); + WARN_ON(1); + return -EINVAL; + } + + rw->slba = cpu_to_le64(((u64)start_lba_hi << 32) | start_lba_lo); + /* 0base for nlb */ + rw->nlb = cpu_to_le16((u16)(datalength - 1)); + rw->control = cpu_to_le16(control); + + return 0; +} + +static int sssraid_setup_nonio_cmd(struct sssraid_dev *hdev, + struct sssraid_scsi_nonio *scsi_nonio, struct scsi_cmnd *scmd) +{ + scsi_nonio->buffer_len = cpu_to_le32(scsi_bufflen(scmd)); + + switch (scmd->sc_data_direction) { + case DMA_NONE: + scsi_nonio->opcode = SSSRAID_CMD_NONIO_NONE; + break; + case DMA_TO_DEVICE: + scsi_nonio->opcode = SSSRAID_CMD_NONIO_TODEV; + break; + case DMA_FROM_DEVICE: + scsi_nonio->opcode = SSSRAID_CMD_NONIO_FROMDEV; + break; + default: + ioc_err(hdev, "Invalid NON_IO for unsupported data direction: %d\n", + scmd->sc_data_direction); + WARN_ON(1); + return -EINVAL; + } + + return 0; +} + +static int sssraid_setup_ioq_cmd(struct sssraid_dev *hdev, + struct sssraid_ioq_command *ioq_cmd, struct scsi_cmnd *scmd) +{ + memcpy(ioq_cmd->common.cdb, scmd->cmnd, scmd->cmd_len); + ioq_cmd->common.cdb_len = scmd->cmd_len; + + if (sssraid_is_rw_scmd(scmd)) + return sssraid_setup_rw_cmd(hdev, &ioq_cmd->rw, scmd); + else + return sssraid_setup_nonio_cmd(hdev, &ioq_cmd->scsi_nonio, scmd); +} + +static inline void sssraid_init_iod(struct sssraid_iod *iod) +{ + iod->nsge = 0; + iod->npages = -1; + iod->use_sgl = false; + WRITE_ONCE(iod->state, SSSRAID_CMD_IDLE); +} + +int sssraid_io_map_data(struct sssraid_dev *hdev, struct sssraid_iod *iod, + struct scsi_cmnd *scmd, struct sssraid_ioq_command *ioq_cmd) +{ + int retval; + + retval = scsi_dma_map(scmd); + if (unlikely(retval < 0)) + return retval; + iod->nsge = retval; + /* No data to DMA, it may be scsi no-rw command */ + if (unlikely(iod->nsge == 0)) + return 0; + + iod->length = scsi_bufflen(scmd); + iod->sg = scsi_sglist(scmd); + iod->use_sgl = !sssraid_is_prp(hdev, scmd, iod->nsge); + + if (iod->use_sgl) { + retval = sssraid_setup_ioq_cmd_sgl(hdev, scmd, ioq_cmd, iod); + } else { + retval = sssraid_setup_prps(hdev, iod); + ioq_cmd->common.dptr.prp1 = + cpu_to_le64(sg_dma_address(iod->sg)); + ioq_cmd->common.dptr.prp2 = cpu_to_le64(iod->first_dma); + } + + if (retval) + scsi_dma_unmap(scmd); + + return retval; +} + +void sssraid_map_status(struct sssraid_iod *iod, struct scsi_cmnd *scmd, + struct sssraid_completion *cqe) +{ + struct sssraid_dev *hdev = iod->sqinfo->hdev; + + scsi_set_resid(scmd, 0); + + switch ((le16_to_cpu(cqe->status) >> 1) & 0x7f) { + case FW_STAT_OK: + set_host_byte(scmd, DID_OK); + break; + case FW_STAT_NEED_CHECK: + set_host_byte(scmd, DID_OK); + scmd->result |= le16_to_cpu(cqe->status) >> 8; + if (scmd->result & SAM_STAT_CHECK_CONDITION) { + memset(scmd->sense_buffer, 0, SCSI_SENSE_BUFFERSIZE); + memcpy(scmd->sense_buffer, iod->sense, SCSI_SENSE_BUFFERSIZE); + scmd->result = (scmd->result & 0x00ffffff) | (DRIVER_SENSE << 24); + } + break; + case FW_STAT_ABORTED: + set_host_byte(scmd, DID_ABORT); + break; + case FW_STAT_NEED_RETRY: + set_host_byte(scmd, DID_REQUEUE); + break; + default: + set_host_byte(scmd, DID_BAD_TARGET); + ioc_warn(hdev, "cid[%d] qid[%d] bad status[0x%x]\n", + le16_to_cpu(cqe->cmd_id), le16_to_cpu(cqe->sq_id), + le16_to_cpu(cqe->status)); + break; + } +} + + +struct sssraid_cmd *sssraid_get_cmd(struct sssraid_dev *hdev, enum sssraid_cmd_type type) +{ + struct sssraid_cmd *cmd = NULL; + unsigned long flags; + struct list_head *head = &hdev->adm_cmd_list; + spinlock_t *slock = &hdev->adm_cmd_lock; + + if (type == SSSRAID_CMD_IOPT) { + head = &hdev->ioq_pt_list; + slock = &hdev->ioq_pt_lock; + } + + spin_lock_irqsave(slock, flags); + if (list_empty(head)) { + spin_unlock_irqrestore(slock, flags); + ioc_err(hdev, "err, cmd[%d] list empty\n", type); + return NULL; + } + cmd = list_entry(head->next, struct sssraid_cmd, list); + list_del_init(&cmd->list); + spin_unlock_irqrestore(slock, flags); + + WRITE_ONCE(cmd->state, SSSRAID_CMD_IN_FLIGHT); + + return cmd; +} + +static int sssraid_add_device(struct sssraid_dev *hdev, struct sssraid_dev_info *device) +{ + struct Scsi_Host *shost = hdev->shost; + struct scsi_device *sdev; + + ioc_info(hdev, "add device, hdid: %u target: %d, channel: %d, lun: %d, attr[0x%x]\n", + le32_to_cpu(device->hdid), le16_to_cpu(device->target), + device->channel, device->lun, device->attr); + + sdev = scsi_device_lookup(shost, device->channel, le16_to_cpu(device->target), 0); + if (sdev) { + ioc_warn(hdev, "Device is already exist, channel: %d, target_id: %d, lun: %d\n", + device->channel, le16_to_cpu(device->target), 0); + scsi_device_put(sdev); + return -EEXIST; + } + scsi_add_device(shost, device->channel, le16_to_cpu(device->target), 0); + return 0; +} + +static int sssraid_rescan_device(struct sssraid_dev *hdev, struct sssraid_dev_info *device) +{ + struct Scsi_Host *shost = hdev->shost; + struct scsi_device *sdev; + + ioc_info(hdev, "rescan device, hdid: %u target: %d, channel: %d, lun: %d, attr[0x%x]\n", + le32_to_cpu(device->hdid), le16_to_cpu(device->target), + device->channel, device->lun, device->attr); + + sdev = scsi_device_lookup(shost, device->channel, le16_to_cpu(device->target), 0); + if (!sdev) { + ioc_warn(hdev, "device is not exit rescan it, channel: %d, target_id: %d, lun: %d\n", + device->channel, le16_to_cpu(device->target), 0); + return -ENODEV; + } + + scsi_rescan_device(&sdev->sdev_gendev); + scsi_device_put(sdev); + return 0; +} + +static int sssraid_remove_device(struct sssraid_dev *hdev, struct sssraid_dev_info *org_device) +{ + struct Scsi_Host *shost = hdev->shost; + struct scsi_device *sdev; + + ioc_info(hdev, "remove device, hdid: %u target: %d, channel: %d, lun: %d, attr[0x%x]\n", + le32_to_cpu(org_device->hdid), le16_to_cpu(org_device->target), + org_device->channel, org_device->lun, org_device->attr); + + sdev = scsi_device_lookup(shost, org_device->channel, le16_to_cpu(org_device->target), 0); + if (!sdev) { + ioc_warn(hdev, "device is not exit remove it, channel: %d, target_id: %d, lun: %d\n", + org_device->channel, le16_to_cpu(org_device->target), 0); + return -ENODEV; + } + + scsi_remove_device(sdev); + scsi_device_put(sdev); + return 0; +} + +static int luntarget_cmp_func(const void *l, const void *r) +{ + const struct sssraid_dev_info *ln = l; + const struct sssraid_dev_info *rn = r; + int l_attr = SSSRAID_DEV_INFO_ATTR_BOOT(ln->attr); + int r_attr = SSSRAID_DEV_INFO_ATTR_BOOT(rn->attr); + + /* boot first */ + if (l_attr != r_attr) + return (r_attr - l_attr); + + if (ln->channel == rn->channel) + return le16_to_cpu(ln->target) - le16_to_cpu(rn->target); + + return ln->channel - rn->channel; +} + +void sssraid_scan_disk(struct sssraid_dev *hdev) +{ + struct sssraid_dev_info *devices, *org_devices; + struct sssraid_dev_info *sortdevice; + u32 nd = le32_to_cpu(hdev->ctrl_info->nd); + u8 flag, org_flag; + int i, ret; + int count = 0; + + devices = kcalloc(nd, sizeof(struct sssraid_dev_info), GFP_KERNEL); + if (!devices) + return; + + sortdevice = kcalloc(nd, sizeof(struct sssraid_dev_info), GFP_KERNEL); + if (!sortdevice) + goto free_list; + + ret = sssraid_get_dev_list(hdev, devices); + if (ret) + goto free_all; + org_devices = hdev->devices; + for (i = 0; i < nd; i++) { + org_flag = org_devices[i].flag; + flag = devices[i].flag; + + dbgprint(hdev, "i: %d, org_flag: 0x%x, flag: 0x%x\n", i, org_flag, flag); + + if (SSSRAID_DEV_INFO_FLAG_VALID(flag)) { + if (!SSSRAID_DEV_INFO_FLAG_VALID(org_flag)) { + down_write(&hdev->devices_rwsem); + memcpy(&org_devices[i], &devices[i], + sizeof(struct sssraid_dev_info)); + memcpy(&sortdevice[count++], &devices[i], + sizeof(struct sssraid_dev_info)); + up_write(&hdev->devices_rwsem); + } else if (SSSRAID_DEV_INFO_FLAG_CHANGE(flag)) { + sssraid_rescan_device(hdev, &devices[i]); + } + } else { + if (SSSRAID_DEV_INFO_FLAG_VALID(org_flag)) { + down_write(&hdev->devices_rwsem); + org_devices[i].flag &= 0xfe; + up_write(&hdev->devices_rwsem); + sssraid_remove_device(hdev, &org_devices[i]); + } + } + } + + ioc_info(hdev, "scan work add device count = %d\n", count); + + sort(sortdevice, count, sizeof(sortdevice[0]), luntarget_cmp_func, NULL); + + for (i = 0; i < count; i++) + sssraid_add_device(hdev, &sortdevice[i]); + +free_all: + kfree(sortdevice); +free_list: + kfree(devices); +} + +static int sssraid_wait_abnl_cmd_done(struct sssraid_iod *iod) +{ + u16 times = 0; + + do { + if (READ_ONCE(iod->state) == SSSRAID_CMD_TMO_COMPLETE) + break; + msleep(500); + times++; + } while (times <= SSSRAID_WAIT_ABNL_CMD_TIMEOUT); + + /* wait command completion timeout after abort/reset success */ + if (times >= SSSRAID_WAIT_ABNL_CMD_TIMEOUT) + return -ETIMEDOUT; + + return 0; +} + +static bool sssraid_check_scmd_completed(struct scsi_cmnd *scmd) +{ + struct sssraid_dev *hdev = shost_priv(scmd->device->host); + struct sssraid_iod *iod = scsi_cmd_priv(scmd); + struct sssraid_squeue *sqinfo; + u16 hwq, cid; + + sssraid_get_tag_from_scmd(scmd, &hwq, &cid); + sqinfo = &hdev->sqinfo[hwq]; + if (READ_ONCE(iod->state) == SSSRAID_CMD_COMPLETE || sssraid_poll_cq(hdev, hwq, cid)) { + ioc_warn(hdev, "cid[%d] qidx[%d] has been completed\n", + cid, sqinfo->qidx); + return true; + } + return false; +} + +static int sssraid_scsi_reset(struct scsi_cmnd *scmd, enum sssraid_scsi_rst_type rst) +{ + struct sssraid_dev *hdev = shost_priv(scmd->device->host); + struct sssraid_iod *iod = scsi_cmd_priv(scmd); + struct sssraid_sdev_hostdata *hostdata; + u16 hwq, cid; + int ret; + + scsi_print_command(scmd); + + if (hdev->state != SSSRAID_LIVE || !sssraid_wait_abnl_cmd_done(iod) || + sssraid_check_scmd_completed(scmd)) + return SUCCESS; + + hostdata = scmd->device->hostdata; + sssraid_get_tag_from_scmd(scmd, &hwq, &cid); + + ioc_warn(hdev, "cid[%d] qidx[%d] timeout, %s reset\n", cid, hwq, + rst ? "bus" : "target"); + ret = sssraid_send_reset_cmd(hdev, rst, hostdata->hdid); + if (ret == 0) { + ret = sssraid_wait_abnl_cmd_done(iod); + if (ret) { + ioc_warn(hdev, "cid[%d] qidx[%d] %s reset failed, not found\n", + cid, hwq, rst ? "bus" : "target"); + return FAILED; + } + + ioc_warn(hdev, "cid[%d] qidx[%d] %s reset success\n", cid, hwq, + rst ? "bus" : "target"); + return SUCCESS; + } + + ioc_warn(hdev, "cid[%d] qidx[%d] ret[%d] %s reset failed\n", cid, hwq, ret, + rst ? "bus" : "target"); + return FAILED; +} + +bool sssraid_change_host_state(struct sssraid_dev *hdev, enum sssraid_state newstate) +{ + unsigned long flags; + enum sssraid_state oldstate; + bool change = false; + + spin_lock_irqsave(&hdev->state_lock, flags); + + oldstate = hdev->state; + switch (newstate) { + case SSSRAID_LIVE: + switch (oldstate) { + case SSSRAID_NEW: + case SSSRAID_RESETTING: + change = true; + break; + default: + break; + } + break; + case SSSRAID_RESETTING: + switch (oldstate) { + case SSSRAID_LIVE: + change = true; + break; + default: + break; + } + break; + case SSSRAID_DELETING: + if (oldstate != SSSRAID_DELETING) + change = true; + break; + case SSSRAID_DEAD: + switch (oldstate) { + case SSSRAID_NEW: + case SSSRAID_LIVE: + case SSSRAID_RESETTING: + change = true; + break; + default: + break; + } + break; + default: + break; + } + if (change) + hdev->state = newstate; + spin_unlock_irqrestore(&hdev->state_lock, flags); + + ioc_info(hdev, "[%d]->[%d], change[%d]\n", oldstate, newstate, change); + + return change; +} + +static int sssraid_get_qd_by_disk(u8 attr) +{ + switch (SSSRAID_DEV_DISK_TYPE(attr)) { + case SSSRAID_SAS_HDD_VD: + case SSSRAID_SATA_HDD_VD: + return SSSRAID_HDD_VD_QD; + case SSSRAID_SAS_SSD_VD: + case SSSRAID_SATA_SSD_VD: + case SSSRAID_NVME_SSD_VD: + return SSSRAID_SSD_VD_QD; + case SSSRAID_SAS_HDD_PD: + case SSSRAID_SATA_HDD_PD: + return SSSRAID_HDD_PD_QD; + case SSSRAID_SAS_SSD_PD: + case SSSRAID_SATA_SSD_PD: + case SSSRAID_NVME_SSD_PD: + return SSSRAID_SSD_PD_QD; + default: + return MAX_CMD_PER_DEV; + } +} + +static int sssraid_match_dev(struct sssraid_dev *hdev, u16 idx, struct scsi_device *sdev) +{ + if (SSSRAID_DEV_INFO_FLAG_VALID(hdev->devices[idx].flag)) { + if (sdev->channel == hdev->devices[idx].channel && + sdev->id == le16_to_cpu(hdev->devices[idx].target) && + sdev->lun < hdev->devices[idx].lun) { + ioc_info(hdev, "Match device success, channel:target:lun[%d:%d:%d]\n", + hdev->devices[idx].channel, + hdev->devices[idx].target, + hdev->devices[idx].lun); + return 1; + } + } + + return 0; +} + +static int sssraid_bsg_map_data(struct sssraid_dev *hdev, struct bsg_job *job, + struct sssraid_admin_command *cmd) +{ + struct request *rq = blk_mq_rq_from_pdu(job); + struct sssraid_iod *iod = job->dd_data; + enum dma_data_direction dma_dir = rq_data_dir(rq) ? DMA_TO_DEVICE : DMA_FROM_DEVICE; + int ret = 0; + + iod->sg = job->request_payload.sg_list; + iod->nsge = job->request_payload.sg_cnt; + iod->length = job->request_payload.payload_len; + iod->use_sgl = false; + iod->npages = -1; + + if (!iod->nsge) + goto out; + + ret = dma_map_sg_attrs(&hdev->pdev->dev, iod->sg, iod->nsge, dma_dir, DMA_ATTR_NO_WARN); + if (!ret) + goto out; + + ret = sssraid_setup_prps(hdev, iod); + if (ret) + goto unmap; + + cmd->common.dptr.prp1 = cpu_to_le64(sg_dma_address(iod->sg)); + cmd->common.dptr.prp2 = cpu_to_le64(iod->first_dma); + + return 0; + +unmap: + dma_unmap_sg(&hdev->pdev->dev, iod->sg, iod->nsge, dma_dir); +out: + return ret; +} + +static void sssraid_bsg_unmap_data(struct sssraid_dev *hdev, struct bsg_job *job) +{ + struct request *rq = blk_mq_rq_from_pdu(job); + struct sssraid_iod *iod = job->dd_data; + enum dma_data_direction dma_dir = rq_data_dir(rq) ? DMA_TO_DEVICE : DMA_FROM_DEVICE; + + if (iod->nsge) + dma_unmap_sg(&hdev->pdev->dev, iod->sg, iod->nsge, dma_dir); + + sssraid_free_iod_res(hdev, iod); +} + +void sssraid_put_cmd(struct sssraid_dev *hdev, struct sssraid_cmd *cmd, + enum sssraid_cmd_type type) +{ + unsigned long flags; + struct list_head *head = &hdev->adm_cmd_list; + spinlock_t *slock = &hdev->adm_cmd_lock; + + if (type == SSSRAID_CMD_IOPT) { + head = &hdev->ioq_pt_list; + slock = &hdev->ioq_pt_lock; + } + + spin_lock_irqsave(slock, flags); + WRITE_ONCE(cmd->state, SSSRAID_CMD_IDLE); + list_add_tail(&cmd->list, head); + spin_unlock_irqrestore(slock, flags); +} + +static int sssraid_user_admin_cmd(struct sssraid_dev *hdev, struct bsg_job *job) +{ + struct sssraid_bsg_request *bsg_req = job->request; + struct sssraid_passthru_common_cmd *cmd = &(bsg_req->admcmd); + struct sssraid_admin_command admin_cmd; + u32 timeout = msecs_to_jiffies(cmd->timeout_ms); + u32 result[2] = {0}; + int status; + + if (hdev->state >= SSSRAID_RESETTING) { + ioc_err(hdev, "err, host state:[%d] is not right\n", + hdev->state); + return -EBUSY; + } + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.common.opcode = cmd->opcode; + admin_cmd.common.flags = cmd->flags; + admin_cmd.common.hdid = cpu_to_le32(cmd->nsid); + admin_cmd.common.cdw2[0] = cpu_to_le32(cmd->cdw2); + admin_cmd.common.cdw2[1] = cpu_to_le32(cmd->cdw3); + admin_cmd.common.cdw10 = cpu_to_le32(cmd->cdw10); + admin_cmd.common.cdw11 = cpu_to_le32(cmd->cdw11); + admin_cmd.common.cdw12 = cpu_to_le32(cmd->cdw12); + admin_cmd.common.cdw13 = cpu_to_le32(cmd->cdw13); + admin_cmd.common.cdw14 = cpu_to_le32(cmd->cdw14); + admin_cmd.common.cdw15 = cpu_to_le32(cmd->cdw15); + + status = sssraid_bsg_map_data(hdev, job, &admin_cmd); + if (status) { + ioc_err(hdev, "err, map data failed\n"); + return status; + } + + status = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, &result[0], &result[1], timeout); + if (status >= 0) { + job->reply_len = sizeof(result); + memcpy(job->reply, result, sizeof(result)); + } + + if (status) + ioc_info(hdev, "opcode[0x%x] subopcode[0x%x], status[0x%x] result0[0x%x] result1[0x%x]\n", + cmd->opcode, cmd->info_0.subopcode, status, result[0], result[1]); + + sssraid_bsg_unmap_data(hdev, job); + + return status; +} + +static int sssraid_submit_ioq_sync_cmd(struct sssraid_dev *hdev, struct sssraid_ioq_command *cmd, + u32 *result, u32 *reslen, u32 timeout) +{ + int ret; + dma_addr_t sense_dma; + struct sssraid_squeue *sqinfo; + void *sense_addr = NULL; + struct sssraid_cmd *pt_cmd = sssraid_get_cmd(hdev, SSSRAID_CMD_IOPT); + + if (!pt_cmd) { + ioc_err(hdev, "err, get sqinfo cmd failed\n"); + return -EFAULT; + } + + timeout = timeout ? timeout : ADMIN_TIMEOUT; + + init_completion(&pt_cmd->cmd_done); + + sqinfo = &hdev->sqinfo[pt_cmd->qid]; + ret = pt_cmd->cid * SCSI_SENSE_BUFFERSIZE; + sense_addr = sqinfo->sense + ret; + sense_dma = sqinfo->sense_dma_addr + ret; + + cmd->common.sense_addr = cpu_to_le64(sense_dma); + cmd->common.sense_len = cpu_to_le16(SCSI_SENSE_BUFFERSIZE); + cmd->common.command_id = cpu_to_le16(pt_cmd->cid); + + sssraid_submit_cmd(sqinfo, cmd); + + if (!wait_for_completion_timeout(&pt_cmd->cmd_done, timeout)) { + ioc_err(hdev, "cid[%d] qid[%d] timeout, opcode[0x%x] subopcode[0x%x]\n", + pt_cmd->cid, pt_cmd->qid, cmd->common.opcode, + (le32_to_cpu(cmd->common.cdw3[0]) & 0xffff)); + + /* reset controller if admin timeout */ + sssraid_adm_timeout(hdev, pt_cmd); + + sssraid_put_cmd(hdev, pt_cmd, SSSRAID_CMD_IOPT); + return -ETIME; + } + + if (result && reslen) { + if ((pt_cmd->status & 0x17f) == 0x101) { + memcpy(result, sense_addr, SCSI_SENSE_BUFFERSIZE); + *reslen = SCSI_SENSE_BUFFERSIZE; + } + } + + sssraid_put_cmd(hdev, pt_cmd, SSSRAID_CMD_IOPT); + + return pt_cmd->status; +} + +static int sssraid_user_ioq_cmd(struct sssraid_dev *hdev, struct bsg_job *job) +{ + struct sssraid_bsg_request *bsg_req = (struct sssraid_bsg_request *)(job->request); + struct sssraid_ioq_passthru_cmd *cmd = &(bsg_req->ioqcmd); + struct sssraid_ioq_command ioq_cmd; + int status = 0; + u32 timeout = msecs_to_jiffies(cmd->timeout_ms); + + if (cmd->data_len > IOQ_PT_DATA_LEN) { + ioc_err(hdev, "data len bigger than 4k\n"); + return -EFAULT; + } + + if (hdev->state != SSSRAID_LIVE) { + ioc_err(hdev, "err, host state:[%d] is not live\n", + hdev->state); + return -EBUSY; + } + + ioc_info(hdev, "opcode[0x%x] subopcode[0x%x] init, datalen[%d]\n", + cmd->opcode, cmd->info_1.subopcode, cmd->data_len); + + memset(&ioq_cmd, 0, sizeof(ioq_cmd)); + ioq_cmd.common.opcode = cmd->opcode; + ioq_cmd.common.flags = cmd->flags; + ioq_cmd.common.hdid = cpu_to_le32(cmd->nsid); + ioq_cmd.common.sense_len = cpu_to_le16(cmd->info_0.res_sense_len); + ioq_cmd.common.cdb_len = cmd->info_0.cdb_len; + ioq_cmd.common.rsvd2 = cmd->info_0.rsvd0; + ioq_cmd.common.cdw3[0] = cpu_to_le32(cmd->cdw3); + ioq_cmd.common.cdw3[1] = cpu_to_le32(cmd->cdw4); + ioq_cmd.common.cdw3[2] = cpu_to_le32(cmd->cdw5); + + ioq_cmd.common.cdw10[0] = cpu_to_le32(cmd->cdw10); + ioq_cmd.common.cdw10[1] = cpu_to_le32(cmd->cdw11); + ioq_cmd.common.cdw10[2] = cpu_to_le32(cmd->cdw12); + ioq_cmd.common.cdw10[3] = cpu_to_le32(cmd->cdw13); + ioq_cmd.common.cdw10[4] = cpu_to_le32(cmd->cdw14); + ioq_cmd.common.cdw10[5] = cpu_to_le32(cmd->data_len); + + memcpy(ioq_cmd.common.cdb, &cmd->cdw16, cmd->info_0.cdb_len); + + ioq_cmd.common.cdw26[0] = cpu_to_le32(cmd->cdw26[0]); + ioq_cmd.common.cdw26[1] = cpu_to_le32(cmd->cdw26[1]); + ioq_cmd.common.cdw26[2] = cpu_to_le32(cmd->cdw26[2]); + ioq_cmd.common.cdw26[3] = cpu_to_le32(cmd->cdw26[3]); + + status = sssraid_bsg_map_data(hdev, job, (struct sssraid_admin_command *)&ioq_cmd); + if (status) { + ioc_err(hdev, "err, map data failed\n"); + return status; + } + + status = sssraid_submit_ioq_sync_cmd(hdev, &ioq_cmd, job->reply, &job->reply_len, timeout); + if (status) + ioc_info(hdev, "opcode[0x%x] subopcode[0x%x], status[0x%x], reply_len[%d]\n", + cmd->opcode, cmd->info_1.subopcode, status, job->reply_len); + + sssraid_bsg_unmap_data(hdev, job); + + return status; +} + + +/* bsg dispatch user command */ +static int sssraid_bsg_host_dispatch(struct bsg_job *job) +{ + struct Scsi_Host *shost = dev_to_shost(job->dev); + struct sssraid_dev *hdev = shost_priv(shost); + struct request *rq = blk_mq_rq_from_pdu(job); + struct sssraid_bsg_request *bsg_req = job->request; + int ret = -ENOMSG; + + job->reply_len = 0; + + if (bsg_req == NULL || job->request_len != sizeof(struct sssraid_bsg_request)) { + bsg_job_done(job, ret, 0); + return 0; + } + dbgprint(hdev, "bsg msgcode[%d] msglen[%d] timeout[%d];" + "reqnsge[%d], reqlen[%d]\n", + bsg_req->msgcode, job->request_len, rq->timeout, + job->request_payload.sg_cnt, job->request_payload.payload_len); + + switch (bsg_req->msgcode) { + case SSSRAID_BSG_ADM: + ret = sssraid_user_admin_cmd(hdev, job); + break; + case SSSRAID_BSG_IOQ: + ret = sssraid_user_ioq_cmd(hdev, job); + break; + default: + ioc_info(hdev, "unsupport msgcode[%d]\n", bsg_req->msgcode); + break; + } + + if (ret > 0) + ret = ret | (ret << 8); + + bsg_job_done(job, ret, 0); + return 0; +} + +static inline void sssraid_remove_bsg(struct sssraid_dev *hdev) +{ + if (hdev->bsg_queue) { + bsg_unregister_queue(hdev->bsg_queue); + blk_cleanup_queue(hdev->bsg_queue); + } +} + +static void sssraid_back_fault_cqe(struct sssraid_squeue *sqinfo, struct sssraid_completion *cqe) +{ + struct sssraid_dev *hdev = sqinfo->hdev; + struct blk_mq_tags *tags; + struct scsi_cmnd *scmd; + struct sssraid_iod *iod; + struct request *req; + + tags = hdev->shost->tag_set.tags[sqinfo->qidx - 1]; + req = blk_mq_tag_to_rq(tags, le16_to_cpu(cqe->cmd_id)); + if (unlikely(!req || !blk_mq_request_started(req))) + return; + + scmd = blk_mq_rq_to_pdu(req); + iod = scsi_cmd_priv(scmd); + + if (READ_ONCE(iod->state) != SSSRAID_CMD_IN_FLIGHT && + READ_ONCE(iod->state) != SSSRAID_CMD_TIMEOUT) + return; + + WRITE_ONCE(iod->state, SSSRAID_CMD_TMO_COMPLETE); + set_host_byte(scmd, DID_NO_CONNECT); + if (iod->nsge) + scsi_dma_unmap(scmd); + sssraid_free_iod_res(hdev, iod); + scmd->scsi_done(scmd); + ioc_warn(hdev, "Back fault CQE, cid[%d] qidx[%d]\n", + le16_to_cpu(cqe->cmd_id), sqinfo->qidx); +} + +void sssraid_back_all_io(struct sssraid_dev *hdev) +{ + int i, j; + struct sssraid_squeue *sqinfo; + struct sssraid_completion cqe = { 0 }; + + scsi_block_requests(hdev->shost); + + for (i = 1; i <= hdev->shost->nr_hw_queues; i++) { + sqinfo = &hdev->sqinfo[i]; + for (j = 0; j < hdev->scsi_qd; j++) { + cqe.cmd_id = cpu_to_le16(j); + sssraid_back_fault_cqe(sqinfo, &cqe); + } + } + + scsi_unblock_requests(hdev->shost); + + j = SSSRAID_AQ_BLK_MQ_DEPTH; + for (i = 0; i < j; i++) { + if (READ_ONCE(hdev->adm_cmds[i].state) == SSSRAID_CMD_IN_FLIGHT) { + ioc_info(hdev, "Back adm, cid[%d]\n", i); + hdev->adm_cmds[i].status = 0xFFFF; + WRITE_ONCE(hdev->adm_cmds[i].state, SSSRAID_CMD_COMPLETE); + complete(&(hdev->adm_cmds[i].cmd_done)); + } + } + + j = SSSRAID_NR_IOQ_PTCMDS; + for (i = 0; i < j; i++) { + if (READ_ONCE(hdev->ioq_ptcmds[i].state) == SSSRAID_CMD_IN_FLIGHT) { + hdev->ioq_ptcmds[i].status = 0xFFFF; + WRITE_ONCE(hdev->ioq_ptcmds[i].state, SSSRAID_CMD_COMPLETE); + complete(&(hdev->ioq_ptcmds[i].cmd_done)); + } + } +} + +/** + * static struct scsi_host_template sssraid_driver_template + */ +static int sssraid_scan_finished(struct Scsi_Host *shost, + unsigned long time) +{ + struct sssraid_dev *hdev = shost_priv(shost); + + sssraid_scan_disk(hdev); + + return 1; +} + +/* eh_target_reset_handler call back */ +static int sssraid_eh_target_reset(struct scsi_cmnd *scmd) +{ + return sssraid_scsi_reset(scmd, SSSRAID_RESET_TARGET); +} + +/* eh_bus_reset_handler call back */ +static int sssraid_bus_reset_handler(struct scsi_cmnd *scmd) +{ + return sssraid_scsi_reset(scmd, SSSRAID_RESET_BUS); +} + +/* eh_host_reset_handler call back */ +static int sssraid_eh_host_reset(struct scsi_cmnd *scmd) +{ + u16 hwq, cid; + struct sssraid_dev *hdev = shost_priv(scmd->device->host); + + scsi_print_command(scmd); + if (hdev->state != SSSRAID_LIVE || sssraid_check_scmd_completed(scmd)) + return SUCCESS; + + sssraid_get_tag_from_scmd(scmd, &hwq, &cid); + ioc_warn(hdev, "cid[%d] qidx[%d] host reset\n", cid, hwq); + + /* It's useless: + * old code sssraid_reset_work_sync + * queue_work(reset_work) at first, + * then flush_work to synchronize. + */ + if (!sssraid_change_host_state(hdev, SSSRAID_RESETTING)) { + ioc_info(hdev, "can't change to reset state\n"); + return FAILED; + } + if (sssraid_soft_reset_handler(hdev)) { + ioc_warn(hdev, "cid[%d] qidx[%d] host reset failed\n", cid, hwq); + return FAILED; + } + + ioc_warn(hdev, "cid[%d] qidx[%d] host reset success\n", cid, hwq); + + return SUCCESS; +} + +/* host_reset call back */ +static int sssraid_sysfs_host_reset(struct Scsi_Host *shost, int reset_type) +{ + int ret; + struct sssraid_dev *hdev = shost_priv(shost); + + ioc_info(hdev, "start sysfs host reset cmd\n"); + if (!sssraid_change_host_state(hdev, SSSRAID_RESETTING)) { + ioc_info(hdev, "can't change to reset state\n"); + return -EBUSY; + } + ret = sssraid_soft_reset_handler(hdev); + ioc_info(hdev, "stop sysfs host reset cmd[%d]\n", ret); + + return ret; +} + +/* queuecommand call back */ +static int sssraid_qcmd(struct Scsi_Host *shost, + struct scsi_cmnd *scmd) +{ + struct sssraid_iod *iod = scsi_cmd_priv(scmd); + struct sssraid_dev *hdev = shost_priv(shost); + struct scsi_device *sdev = scmd->device; + struct sssraid_sdev_hostdata *hostdata = sdev->hostdata; + u16 hwq, cid; + struct sssraid_squeue *sq; + struct sssraid_ioq_command ioq_cmd; + int retval; + + if (unlikely(hdev->state == SSSRAID_RESETTING)) + return SCSI_MLQUEUE_HOST_BUSY; + + if (unlikely(hdev->state != SSSRAID_LIVE)) { + set_host_byte(scmd, DID_NO_CONNECT); + scmd->scsi_done(scmd); + return 0; + } + + if (unlikely(hdev->logging_level & SSSRAID_DEBUG)) + scsi_print_command(scmd); + + sssraid_get_tag_from_scmd(scmd, &hwq, &cid); + hostdata = sdev->hostdata; + sq = &hdev->sqinfo[hwq]; + + memset(&ioq_cmd, 0, sizeof(ioq_cmd)); + ioq_cmd.rw.hdid = cpu_to_le32(hostdata->hdid); + ioq_cmd.rw.command_id = cpu_to_le16(cid); + + retval = sssraid_setup_ioq_cmd(hdev, &ioq_cmd, scmd); + if (unlikely(retval)) { + set_host_byte(scmd, DID_ERROR); + scmd->scsi_done(scmd); + return 0; + } + + iod->sense = sq->sense + retval; + iod->sense_dma = sq->sense_dma_addr + retval; + ioq_cmd.common.sense_addr = cpu_to_le64(iod->sense_dma); + ioq_cmd.common.sense_len = cpu_to_le16(SCSI_SENSE_BUFFERSIZE); + + sssraid_init_iod(iod); + + iod->sqinfo = sq; + sssraid_io_map_data(hdev, iod, scmd, &ioq_cmd); + if (unlikely(retval)) { + ioc_err(hdev, "sssraid_io_map_data Err.\n"); + set_host_byte(scmd, DID_ERROR); + scmd->scsi_done(scmd); + retval = 0; + goto deinit_iod; + } + + WRITE_ONCE(iod->state, SSSRAID_CMD_IN_FLIGHT); + sssraid_submit_cmd(sq, &ioq_cmd); + + return 0; + +deinit_iod: + sssraid_free_iod_res(hdev, iod); + return retval; +} + +/* change_queue_depth call back: + * keep as old + */ + +/* slave_configure call back */ +static int sssraid_slave_configure(struct scsi_device *sdev) +{ + int qd = MAX_CMD_PER_DEV; + unsigned int timeout = scmd_tmout_rawdisk * HZ; + struct sssraid_dev *hdev = shost_priv(sdev->host); + struct sssraid_sdev_hostdata *hostdata = sdev->hostdata; + u32 max_sec = sdev->host->max_sectors; + + if (hostdata) { + if (SSSRAID_DEV_INFO_ATTR_VD(hostdata->attr)) + timeout = scmd_tmout_vd * HZ; + else if (SSSRAID_DEV_INFO_ATTR_RAWDISK(hostdata->attr)) + timeout = scmd_tmout_rawdisk * HZ; + max_sec = hostdata->max_io_kb << 1; + qd = sssraid_get_qd_by_disk(hostdata->attr); + } else { + ioc_err(hdev, "err, sdev->hostdata is null\n"); + } + + blk_queue_rq_timeout(sdev->request_queue, timeout); + sdev->eh_timeout = timeout; + scsi_change_queue_depth(sdev, qd); + + if ((max_sec == 0) || (max_sec > sdev->host->max_sectors)) + max_sec = sdev->host->max_sectors; + + blk_queue_max_hw_sectors(sdev->request_queue, max_sec); + + ioc_info(hdev, "sdev->channel:id:lun[%d:%d:%lld], scmd_timeout[%d]s, maxsec[%d]\n", + sdev->channel, sdev->id, sdev->lun, timeout / HZ, max_sec); + + return 0; +} + +/* slave_alloc call back */ +static int sssraid_slave_alloc(struct scsi_device *sdev) +{ + struct sssraid_sdev_hostdata *hostdata; + struct sssraid_dev *hdev; + u16 idx; + + hdev = shost_priv(sdev->host); + hostdata = kzalloc(sizeof(*hostdata), GFP_KERNEL); + if (!hostdata) { + ioc_err(hdev, "Alloc scsi host data memory failed\n"); + return -ENOMEM; + } + + down_read(&hdev->devices_rwsem); + for (idx = 0; idx < le32_to_cpu(hdev->ctrl_info->nd); idx++) { + if (sssraid_match_dev(hdev, idx, sdev)) + goto scan_host; + } + up_read(&hdev->devices_rwsem); + + kfree(hostdata); + return -ENXIO; + +scan_host: + hostdata->hdid = le32_to_cpu(hdev->devices[idx].hdid); + hostdata->max_io_kb = le16_to_cpu(hdev->devices[idx].max_io_kb); + hostdata->attr = hdev->devices[idx].attr; + hostdata->flag = hdev->devices[idx].flag; + hostdata->rg_id = 0xff; + sdev->hostdata = hostdata; + up_read(&hdev->devices_rwsem); + return 0; +} + +/* slave_destroy call back */ +static void sssraid_slave_destroy(struct scsi_device *sdev) +{ + kfree(sdev->hostdata); + sdev->hostdata = NULL; +} + +/* eh_timed_out call back */ +static enum blk_eh_timer_return sssraid_scmd_timeout(struct scsi_cmnd *scmd) +{ + struct sssraid_iod *iod = scsi_cmd_priv(scmd); + unsigned int timeout = scmd->device->request_queue->rq_timeout; + + if (sssraid_check_scmd_completed(scmd)) + goto out; + + if (time_after(jiffies, scmd->jiffies_at_alloc + timeout)) { + if (cmpxchg(&iod->state, SSSRAID_CMD_IN_FLIGHT, SSSRAID_CMD_TIMEOUT) == + SSSRAID_CMD_IN_FLIGHT) { + return BLK_EH_DONE; + } + } +out: + return BLK_EH_RESET_TIMER; +} + +/* eh_abort_handler call back */ +static int sssraid_abort_handler(struct scsi_cmnd *scmd) +{ + struct sssraid_dev *hdev = shost_priv(scmd->device->host); + struct sssraid_iod *iod = scsi_cmd_priv(scmd); + struct sssraid_sdev_hostdata *hostdata; + u16 hwq, cid; + int ret; + + scsi_print_command(scmd); + + if (hdev->state != SSSRAID_LIVE || !sssraid_wait_abnl_cmd_done(iod) || + sssraid_check_scmd_completed(scmd)) + return SUCCESS; + + hostdata = scmd->device->hostdata; + sssraid_get_tag_from_scmd(scmd, &hwq, &cid); + + ioc_warn(hdev, "cid[%d] qidx[%d] timeout, aborting\n", cid, hwq); + ret = sssraid_send_abort_cmd(hdev, hostdata->hdid, hwq, cid); + if (ret != -ETIME) { + ret = sssraid_wait_abnl_cmd_done(iod); + if (ret) { + ioc_warn(hdev, "cid[%d] qidx[%d] abort failed, not found\n", cid, hwq); + return FAILED; + } + ioc_warn(hdev, "cid[%d] qidx[%d] abort succ\n", cid, hwq); + return SUCCESS; + } + ioc_warn(hdev, "cid[%d] qidx[%d] abort failed, timeout\n", cid, hwq); + return FAILED; +} + +static ssize_t csts_pp_show(struct device *cdev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct sssraid_dev *hdev = shost_priv(shost); + int ret = -1; + + if (pci_device_is_present(hdev->pdev)) { + ret = (readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_PP_MASK); + ret >>= SSSRAID_CSTS_PP_SHIFT; + } + + return snprintf(buf, PAGE_SIZE, "%d\n", ret); +} + +static ssize_t csts_shst_show(struct device *cdev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct sssraid_dev *hdev = shost_priv(shost); + int ret = -1; + + if (pci_device_is_present(hdev->pdev)) { + ret = (readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_SHST_MASK); + ret >>= SSSRAID_CSTS_SHST_SHIFT; + } + + return snprintf(buf, PAGE_SIZE, "%d\n", ret); +} + +static ssize_t csts_cfs_show(struct device *cdev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct sssraid_dev *hdev = shost_priv(shost); + int ret = -1; + + if (pci_device_is_present(hdev->pdev)) { + ret = (readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_CFS_MASK); + ret >>= SSSRAID_CSTS_CFS_SHIFT; + } + + return snprintf(buf, PAGE_SIZE, "%d\n", ret); +} + +static ssize_t csts_rdy_show(struct device *cdev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct sssraid_dev *hdev = shost_priv(shost); + int ret = -1; + + if (pci_device_is_present(hdev->pdev)) + ret = (readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_RDY); + + return snprintf(buf, PAGE_SIZE, "%d\n", ret); +} + +static ssize_t fw_version_show(struct device *cdev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct sssraid_dev *hdev = shost_priv(shost); + + return snprintf(buf, PAGE_SIZE, "%s\n", hdev->ctrl_info->fr); +} + +static DEVICE_ATTR_RO(csts_pp); +static DEVICE_ATTR_RO(csts_shst); +static DEVICE_ATTR_RO(csts_cfs); +static DEVICE_ATTR_RO(csts_rdy); +static DEVICE_ATTR_RO(fw_version); + +static struct device_attribute *sssraid_host_attrs[] = { + &dev_attr_csts_pp, + &dev_attr_csts_shst, + &dev_attr_csts_cfs, + &dev_attr_csts_rdy, + &dev_attr_fw_version, + NULL, +}; + +static int sssraid_get_vd_info(struct sssraid_dev *hdev, struct sssraid_vd_info *vd_info, u16 vid) +{ + struct sssraid_admin_command admin_cmd; + u8 *data_ptr = NULL; + dma_addr_t data_dma = 0; + int ret; + + if (hdev->state >= SSSRAID_RESETTING) { + ioc_err(hdev, "err, host state[%d] is not right\n", hdev->state); + return -EBUSY; + } + + data_ptr = dma_alloc_coherent(&hdev->pdev->dev, PAGE_SIZE, &data_dma, GFP_KERNEL); + if (!data_ptr) + return -ENOMEM; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.usr_cmd.opcode = USR_CMD_READ; + admin_cmd.usr_cmd.info_0.subopcode = cpu_to_le16(USR_CMD_VDINFO); + admin_cmd.usr_cmd.info_1.data_len = cpu_to_le16(USR_CMD_RDLEN); + admin_cmd.usr_cmd.info_1.param_len = cpu_to_le16(VDINFO_PARAM_LEN); + admin_cmd.usr_cmd.cdw10 = cpu_to_le32(vid); + admin_cmd.common.dptr.prp1 = cpu_to_le64(data_dma); + + ret = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); + if (!ret) + memcpy(vd_info, data_ptr, sizeof(struct sssraid_vd_info)); + + dma_free_coherent(&hdev->pdev->dev, PAGE_SIZE, data_ptr, data_dma); + + return ret; +} + +static int sssraid_get_bgtask(struct sssraid_dev *hdev, struct sssraid_bgtask *bgtask) +{ + struct sssraid_admin_command admin_cmd; + u8 *data_ptr = NULL; + dma_addr_t data_dma = 0; + int ret; + + if (hdev->state >= SSSRAID_RESETTING) { + ioc_err(hdev, "err, host state[%d] is not right\n", hdev->state); + return -EBUSY; + } + + data_ptr = dma_alloc_coherent(&hdev->pdev->dev, PAGE_SIZE, &data_dma, GFP_KERNEL); + if (!data_ptr) + return -ENOMEM; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.usr_cmd.opcode = USR_CMD_READ; + admin_cmd.usr_cmd.info_0.subopcode = cpu_to_le16(USR_CMD_BGTASK); + admin_cmd.usr_cmd.info_1.data_len = cpu_to_le16(USR_CMD_RDLEN); + admin_cmd.common.dptr.prp1 = cpu_to_le64(data_dma); + + ret = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); + if (!ret) + memcpy(bgtask, data_ptr, sizeof(struct sssraid_bgtask)); + + dma_free_coherent(&hdev->pdev->dev, PAGE_SIZE, data_ptr, data_dma); + + return ret; +} + +static ssize_t raid_level_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct scsi_device *sdev; + struct sssraid_dev *hdev; + struct sssraid_vd_info *vd_info; + struct sssraid_sdev_hostdata *hostdata; + int ret; + + sdev = to_scsi_device(dev); + hdev = shost_priv(sdev->host); + hostdata = sdev->hostdata; + + vd_info = kmalloc(sizeof(*vd_info), GFP_KERNEL); + if (!vd_info || !SSSRAID_DEV_INFO_ATTR_VD(hostdata->attr)) + return snprintf(buf, PAGE_SIZE, "NA\n"); + + ret = sssraid_get_vd_info(hdev, vd_info, sdev->id); + if (ret) + vd_info->rg_level = ARRAY_SIZE(raid_levels) - 1; + + ret = (vd_info->rg_level < ARRAY_SIZE(raid_levels)) ? + vd_info->rg_level : (ARRAY_SIZE(raid_levels) - 1); + + kfree(vd_info); + + return snprintf(buf, PAGE_SIZE, "RAID-%s\n", raid_levels[ret]); +} + +static ssize_t raid_state_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct scsi_device *sdev; + struct sssraid_dev *hdev; + struct sssraid_vd_info *vd_info; + struct sssraid_sdev_hostdata *hostdata; + int ret; + + sdev = to_scsi_device(dev); + hdev = shost_priv(sdev->host); + hostdata = sdev->hostdata; + + vd_info = kmalloc(sizeof(*vd_info), GFP_KERNEL); + if (!vd_info || !SSSRAID_DEV_INFO_ATTR_VD(hostdata->attr)) + return snprintf(buf, PAGE_SIZE, "NA\n"); + + ret = sssraid_get_vd_info(hdev, vd_info, sdev->id); + if (ret) { + vd_info->vd_status = 0; + vd_info->rg_id = 0xff; + } + + ret = (vd_info->vd_status < ARRAY_SIZE(raid_states)) ? vd_info->vd_status : 0; + + kfree(vd_info); + + return snprintf(buf, PAGE_SIZE, "%s\n", raid_states[ret]); +} + +static ssize_t raid_resync_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct scsi_device *sdev; + struct sssraid_dev *hdev; + struct sssraid_vd_info *vd_info; + struct sssraid_bgtask *bgtask; + struct sssraid_sdev_hostdata *hostdata; + u8 rg_id, i, progress = 0; + int ret; + + sdev = to_scsi_device(dev); + hdev = shost_priv(sdev->host); + hostdata = sdev->hostdata; + + vd_info = kmalloc(sizeof(*vd_info), GFP_KERNEL); + if (!vd_info || !SSSRAID_DEV_INFO_ATTR_VD(hostdata->attr)) + return snprintf(buf, PAGE_SIZE, "NA\n"); + + ret = sssraid_get_vd_info(hdev, vd_info, sdev->id); + if (ret) + goto out; + + rg_id = vd_info->rg_id; + + bgtask = (struct sssraid_bgtask *)vd_info; + ret = sssraid_get_bgtask(hdev, bgtask); + if (ret) + goto out; + for (i = 0; i < bgtask->task_num; i++) { + if ((bgtask->bgtask[i].type == BGTASK_TYPE_REBUILD) && + (le16_to_cpu(bgtask->bgtask[i].vd_id) == rg_id)) + progress = bgtask->bgtask[i].progress; + } + +out: + kfree(vd_info); + return snprintf(buf, PAGE_SIZE, "%d\n", progress); +} + +static DEVICE_ATTR_RO(raid_level); +static DEVICE_ATTR_RO(raid_state); +static DEVICE_ATTR_RO(raid_resync); + +static struct device_attribute *sssraid_dev_attrs[] = { + &dev_attr_raid_level, + &dev_attr_raid_state, + &dev_attr_raid_resync, + NULL, +}; + +static struct scsi_host_template sssraid_driver_template = { + .module = THIS_MODULE, + .name = "3SNIC Logic sssraid driver", + .proc_name = "sssraid", + .queuecommand = sssraid_qcmd, + .slave_alloc = sssraid_slave_alloc, + .slave_destroy = sssraid_slave_destroy, + .slave_configure = sssraid_slave_configure, + .scan_finished = sssraid_scan_finished, + .eh_timed_out = sssraid_scmd_timeout, + .eh_abort_handler = sssraid_abort_handler, + .eh_target_reset_handler = sssraid_eh_target_reset, + .eh_bus_reset_handler = sssraid_bus_reset_handler, + .eh_host_reset_handler = sssraid_eh_host_reset, + .change_queue_depth = scsi_change_queue_depth, + .host_tagset = 0, + .this_id = -1, + .unchecked_isa_dma = 0, + .shost_attrs = sssraid_host_attrs, + .sdev_attrs = sssraid_dev_attrs, + .host_reset = sssraid_sysfs_host_reset, +}; + +/** + * sssraid_probe - PCI probe callback + * @pdev: PCI device instance + * @id: PCI device ID details + * + * controller initialization routine. + * Allocate per adapter instance through shost_priv and + * initialize controller specific data structures, initializae + * the controller hardware, add shost to the SCSI subsystem. + * + * Return: 0 on success, non-zero on failure. + */ + +static int +sssraid_probe(struct pci_dev *pdev, const struct pci_device_id *id) +{ + struct sssraid_dev *hdev; + struct Scsi_Host *shost; + int node; + char bsg_name[15]; + int retval = 0; + + node = dev_to_node(&pdev->dev); + if (node == NUMA_NO_NODE) { + node = first_memory_node; + set_dev_node(&pdev->dev, node); + } + + shost = scsi_host_alloc(&sssraid_driver_template, sizeof(*hdev)); + if (!shost) { + retval = -ENODEV; + ioc_err(hdev, "Failed to allocate scsi host\n"); + goto shost_failed; + } + + hdev = shost_priv(shost); + hdev->numa_node = node; + hdev->instance = shost->host_no; /* for device instance */ + sprintf(hdev->name, "%s%d", SSSRAID_DRIVER_NAME, hdev->instance); + + init_rwsem(&hdev->devices_rwsem); + spin_lock_init(&hdev->state_lock); + + spin_lock_init(&hdev->fwevt_lock); + spin_lock_init(&hdev->watchdog_lock); + + INIT_LIST_HEAD(&hdev->fwevt_list); + +// logging_level = 1; //garden test + hdev->logging_level = logging_level; /* according to log_debug_switch*/ + + snprintf(hdev->fwevt_worker_name, sizeof(hdev->fwevt_worker_name), + "%s%d_fwevt_wrkr", SSSRAID_DRIVER_NAME, hdev->instance); + hdev->fwevt_worker_thread = alloc_ordered_workqueue( + hdev->fwevt_worker_name, WQ_MEM_RECLAIM); + if (!hdev->fwevt_worker_thread) { + ioc_err(hdev, "failure at %s:%d/%s()!\n", + __FILE__, __LINE__, __func__); + retval = -ENODEV; + goto out_fwevtthread_failed; + } + + hdev->shost = shost; + hdev->pdev = pdev; + + if (sssraid_init_ioc(hdev, 0)) { + ioc_err(hdev, "failure at %s:%d/%s()!\n", + __FILE__, __LINE__, __func__); + retval = -ENODEV; + goto out_iocinit_failed; + } + + sssraid_shost_init(hdev); + + retval = scsi_add_host(shost, &pdev->dev); + if (retval) { + ioc_err(hdev, "failure at %s:%d/%s()!\n", + __FILE__, __LINE__, __func__); + goto addhost_failed; + } + + snprintf(bsg_name, sizeof(bsg_name), "%s%d", SSSRAID_DRIVER_NAME, shost->host_no); + hdev->bsg_queue = bsg_setup_queue(&shost->shost_gendev, bsg_name, + sssraid_bsg_host_dispatch, NULL, sssraid_cmd_size(hdev)); + if (IS_ERR(hdev->bsg_queue)) { + ioc_err(hdev, "err, setup bsg failed\n"); + hdev->bsg_queue = NULL; + goto bsg_setup_failed; + } + + sssraid_change_host_state(hdev, SSSRAID_LIVE); + + scsi_scan_host(shost); + return retval; + +bsg_setup_failed: + scsi_remove_host(shost); +addhost_failed: + sssraid_cleanup_ioc(hdev, 0); +out_iocinit_failed: + destroy_workqueue(hdev->fwevt_worker_thread); +out_fwevtthread_failed: + scsi_host_put(shost); +shost_failed: + return retval; +} + +static void sssraid_remove(struct pci_dev *pdev) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + + if (!shost) + return; + + ioc_info(hdev, "enter sssraid remove\n"); + + hdev = shost_priv(shost); + + sssraid_change_host_state(hdev, SSSRAID_DELETING); + + if (!pci_device_is_present(pdev)) + sssraid_back_all_io(hdev); + + sssraid_cleanup_fwevt_list(hdev); + destroy_workqueue(hdev->fwevt_worker_thread); + + sssraid_remove_bsg(hdev); + scsi_remove_host(shost); + sssraid_cleanup_ioc(hdev, 0); + + scsi_host_put(shost); +} + +static void sssraid_shutdown(struct pci_dev *pdev) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + + if (!shost) + return; + + hdev = shost_priv(shost); + + sssraid_cleanup_fwevt_list(hdev); + destroy_workqueue(hdev->fwevt_worker_thread); + sssraid_cleanup_ioc(hdev, 0); +} + +#ifdef CONFIG_PM +static int sssraid_suspend(struct pci_dev *pdev, pm_message_t state) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + pci_power_t device_state; + + if (!shost) + return 0; + + hdev = shost_priv(shost); + + while (hdev->state == SSSRAID_RESETTING) + ssleep(1); + sssraid_cleanup_fwevt_list(hdev); + scsi_block_requests(shost); + sssraid_cleanup_ioc(hdev, 1); + + device_state = pci_choose_state(pdev, state); + pci_save_state(pdev); + pci_set_power_state(pdev, device_state); + + return 0; +} + +static int sssraid_resume(struct pci_dev *pdev) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + pci_power_t device_state = pdev->current_state; + + if (!shost) + return 0; + + hdev = shost_priv(shost); + + ioc_info(hdev, "pdev=0x%p, slot=%s, previous operating state [D%d]\n", + pdev, pci_name(pdev), device_state); + pci_set_power_state(pdev, PCI_D0); + pci_enable_wake(pdev, PCI_D0, 0); + pci_restore_state(pdev); + hdev->pdev = pdev; + hdev->cpu_count = num_online_cpus(); + + /* sssraid_setup_resources in sssraid_init_ioc */ + sssraid_init_ioc(hdev, 1); + scsi_unblock_requests(shost); + + return 0; +} +#endif + +static pci_ers_result_t sssraid_pci_error_detected(struct pci_dev *pdev, + pci_channel_state_t state) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + + if (!shost) + return PCI_ERS_RESULT_NONE; + + hdev = shost_priv(shost); + + ioc_info(hdev, "enter pci error detect, state:%d\n", state); + + switch (state) { + case pci_channel_io_normal: + ioc_warn(hdev, "channel is normal, do nothing\n"); + + return PCI_ERS_RESULT_CAN_RECOVER; + case pci_channel_io_frozen: + ioc_warn(hdev, "channel io frozen, need reset controller\n"); + + scsi_block_requests(hdev->shost); + + sssraid_change_host_state(hdev, SSSRAID_RESETTING); + + return PCI_ERS_RESULT_NEED_RESET; + case pci_channel_io_perm_failure: + ioc_warn(hdev, "channel io failure, request disconnect\n"); + + return PCI_ERS_RESULT_DISCONNECT; + } + + return PCI_ERS_RESULT_NEED_RESET; +} + +static pci_ers_result_t sssraid_pci_slot_reset(struct pci_dev *pdev) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + + if (!shost) + return PCI_ERS_RESULT_NONE; + + hdev = shost_priv(shost); + + ioc_info(hdev, "restart after slot reset\n"); + + pci_restore_state(pdev); + + sssraid_soft_reset_handler(hdev); + + scsi_unblock_requests(hdev->shost); + + return PCI_ERS_RESULT_RECOVERED; +} + +static void sssraid_reset_done(struct pci_dev *pdev) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + + if (!shost) + return; + + hdev = shost_priv(shost); + + ioc_info(hdev, "enter sssraid reset done\n"); +} + +static struct pci_error_handlers sssraid_err_handler = { + .error_detected = sssraid_pci_error_detected, + .slot_reset = sssraid_pci_slot_reset, + .reset_done = sssraid_reset_done, +}; + +static const struct pci_device_id sssraid_pci_id_table[] = { + { PCI_DEVICE(PCI_VENDOR_ID_3SNIC_LOGIC, SSSRAID_SERVER_DEVICE_HBA_DID) }, + { PCI_DEVICE(PCI_VENDOR_ID_3SNIC_LOGIC, SSSRAID_SERVER_DEVICE_RAID_DID) }, + { 0, } +}; +MODULE_DEVICE_TABLE(pci, sssraid_pci_id_table); + +static struct pci_driver sssraid_pci_driver = { + .name = SSSRAID_DRIVER_NAME, + .id_table = sssraid_pci_id_table, + .probe = sssraid_probe, + .remove = sssraid_remove, + .shutdown = sssraid_shutdown, +#ifdef CONFIG_PM + .suspend = sssraid_suspend, + .resume = sssraid_resume, +#endif + .err_handler = &sssraid_err_handler, +}; + +static int __init sssraid_init(void) +{ + int ret_val; + + pr_info("Loading %s version %s\n", SSSRAID_DRIVER_NAME, + SSSRAID_DRIVER_VERSION); + + sssraid_class = class_create(THIS_MODULE, "sssraid"); + if (IS_ERR(sssraid_class)) { + ret_val = PTR_ERR(sssraid_class); + return ret_val; + } + + ret_val = pci_register_driver(&sssraid_pci_driver); + + return ret_val; +} + +static void __exit sssraid_exit(void) +{ + pr_info("Unloading %s version %s\n", SSSRAID_DRIVER_NAME, + SSSRAID_DRIVER_VERSION); + + class_destroy(sssraid_class); + pci_unregister_driver(&sssraid_pci_driver); +} + +MODULE_AUTHOR("liangry1(a)3snic.com"); +MODULE_DESCRIPTION("3SNIC Information Technology SSSRAID Driver"); +MODULE_LICENSE("GPL"); +MODULE_VERSION(SSSRAID_DRIVER_VERSION); +module_init(sssraid_init); +module_exit(sssraid_exit); diff --git a/openEuler/MAINTAINERS b/openEuler/MAINTAINERS index 35b5a8f81fed..fc8b6e06fc04 100644 --- a/openEuler/MAINTAINERS +++ b/openEuler/MAINTAINERS @@ -76,6 +76,12 @@ F: include/linux/wait.h F: include/uapi/linux/sched.h F: kernel/sched/ +SSSRAID SCSI/Raid DRIVERS +M: Steven Song <steven.song(a)3snic.com> +S: Maintained +F: Documentation/scsi/sssraid.rst +F: drivers/scsi/sssraid/ + THE REST M: xiexiuqi(a)huawei.com M: zhengzengkai(a)huawei.com -- 2.25.1

1 0

[PATCH] Support 3SNIC 3S5XX serial RAID/HBA controllers
by Steven Song 29 Dec '22

29 Dec '22

3SNIC inclusion category: features bugzilla: https://gitee.com/openeuler/kernel/issues/I64M90 CVE: NA This commit is to suppport 3SNIC 3S5XX RAID/HBA controllers. RAID controllers support RAID 0/1/5/6/10/50/60 modes; HBA controlllers support RAID 0/1/10 modes. RAID/HBA support SAS/SATA HDD/SSD. Signed-off-by: liangry <liangry1(a)3snic.com> Reviewed-by: Jiang Yu<yujiang(a)3snic.com> Reviewed-by: Steven Song <steven.song(a)3snic.com> --- Documentation/scsi/sssraid.rst | 83 + MAINTAINERS | 7 + arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + drivers/scsi/Kconfig | 1 + drivers/scsi/Makefile | 1 + drivers/scsi/sssraid/Kconfig | 14 + drivers/scsi/sssraid/Makefile | 7 + drivers/scsi/sssraid/sssraid.h | 971 ++++++++++ drivers/scsi/sssraid/sssraid_debug.h | 34 + drivers/scsi/sssraid/sssraid_fw.c | 1724 +++++++++++++++++ drivers/scsi/sssraid/sssraid_os.c | 2337 ++++++++++++++++++++++++ 12 files changed, 5181 insertions(+) create mode 100644 Documentation/scsi/sssraid.rst create mode 100644 drivers/scsi/sssraid/Kconfig create mode 100644 drivers/scsi/sssraid/Makefile create mode 100644 drivers/scsi/sssraid/sssraid.h create mode 100644 drivers/scsi/sssraid/sssraid_debug.h create mode 100644 drivers/scsi/sssraid/sssraid_fw.c create mode 100644 drivers/scsi/sssraid/sssraid_os.c diff --git a/Documentation/scsi/sssraid.rst b/Documentation/scsi/sssraid.rst new file mode 100644 index 000000000000..d56e5f7c478c --- /dev/null +++ b/Documentation/scsi/sssraid.rst @@ -0,0 +1,83 @@ +.. SPDX-License-Identifier: GPL-2.0 + +============================================== +SSSRAID - 3SNIC SCSI RAID Controller driver +============================================== + +This file describes the SSSRAID SCSI driver for 3SNIC +(http://www.3snic.com) RAID controllers. The SSSRAID +driver is the first generation RAID driver for 3SNIC Corp. + +For 3SNIC SSSRAID controller support, enable the SSSRAID driver +when configuring the kernel. + +SSSRAID specific entries in /sys +================================= + +SSSRAID host attributes +------------------------ + - /sys/class/scsi_host/host*/csts_pp + - /sys/class/scsi_host/host*/csts_shst + - /sys/class/scsi_host/host*/csts_cfs + - /sys/class/scsi_host/host*/csts_rdy + - /sys/class/scsi_host/host*/fw_version + + The host csts_pp attribute is a read only attribute. This attribute + indicates whether the controller is processing commands. If this attribute + is set to ‘1’, then the controller is processing commands normally. If + this attribute is cleared to ‘0’, then the controller has temporarily stopped + processing commands in order to handle an event (e.g., firmware activation). + + The host csts_shst attribute is a read only attribute. This attribute + indicates status of shutdown processing.The shutdown status values are defined + as: + ====== ============================== + Value Definition + ====== ============================== + 00b Normal operation + 01b Shutdown processing occurring + 10b Shutdown processing complete + 11b Reserved + ====== ============================== + The host csts_cfs attribute is a read only attribute. This attribute is set to + ’1’ when a fatal controller error occurred that could not be communicated in the + appropriate Completion Queue. This bit is cleared to ‘0’ when a fatal controller + error has not occurred. + + The host csts_rdy attribute is a read only attribute. This attribute is set to + ‘1’ when the controller is ready to process submission queue entries. + + The fw_version attribute is read-only and will return the driver version and the + controller firmware version. + +SSSRAID scsi device attributes +------------------------------ + - /sys/class/scsi_device/X\:X\:X\:X/device/raid_level + - /sys/class/scsi_device/X\:X\:X\:X/device/raid_state + - /sys/class/scsi_device/X\:X\:X\:X/device/raid_resync + + The device raid_level attribute is a read only attribute. This attribute indicates + RAID level of scsi device(will dispaly "NA" if scsi device is not virtual disk type). + + The device raid_state attribute is read-only and indicates RAID status of scsi + device(will dispaly "NA" if scsi device is not virtual disk type). + + The device raid_resync attribute is read-only and indicates RAID rebuild processing + of scsi device(will dispaly "NA" if scsi device is not virtual disk type). + +Supported devices +================= + + =================== ======= ======================================= + PCI ID (pci.ids) OEM Product + =================== ======= ======================================= + 1F3F:2100 3SNIC 3S510(HBA:8Ports,1G DDR) + 1F3F:2100 3SNIC 3S520(HBA:16Ports,1G DDR) + 1F3F:2100 3SNIC 3S530(HBA:32Ports,1G DDR) + 1F3F:2100 3SNIC 3S540(HBA:40Ports,1G DDR) + 1F3F:2200 3SNIC 3S580(RAID:16Ports,2G cache) + 1F3F:2200 3SNIC 3S585(RAID:16Ports,4G cache) + 1F3F:2200 3SNIC 3S590(RAID:32Ports,4G cache) + 1F3F:2200 3SNIC 3S5A0(RAID:40Ports,2G cache) + 1F3F:2200 3SNIC 3S5A5(RAID:40Ports,4G cache) + =================== ======= ======================================= diff --git a/MAINTAINERS b/MAINTAINERS index 9ae209633628..e6470640f764 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16669,6 +16669,13 @@ M: Jan-Benedict Glaw <jbglaw(a)lug-owl.de> S: Maintained F: arch/alpha/kernel/srm_env.c +SSSRAID SCSI/Raid DRIVERS +M: Steven Song <steven.song(a)3snic.com> +L: linux-scsi(a)vger.kernel.org +S: Maintained +F: Documentation/scsi/sssraid.rst +F: drivers/scsi/sssraid/ + ST LSM6DSx IMU IIO DRIVER M: Lorenzo Bianconi <lorenzo.bianconi83(a)gmail.com> L: linux-iio(a)vger.kernel.org diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 30384eacad4f..dc4ba0d2b694 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -2394,6 +2394,7 @@ CONFIG_SCSI_HISI_SAS_DEBUGFS_DEFAULT_ENABLE=y # CONFIG_MEGARAID_NEWGEN is not set # CONFIG_MEGARAID_LEGACY is not set CONFIG_MEGARAID_SAS=m +CONFIG_SCSI_3SNIC_SSSRAID=m CONFIG_SCSI_MPT3SAS=m CONFIG_SCSI_MPT2SAS_MAX_SGE=128 CONFIG_SCSI_MPT3SAS_MAX_SGE=128 diff --git a/arch/x86/configs/openeuler_defconfig b/arch/x86/configs/openeuler_defconfig index 5ada612f1d75..df16f123515b 100644 --- a/arch/x86/configs/openeuler_defconfig +++ b/arch/x86/configs/openeuler_defconfig @@ -2355,6 +2355,7 @@ CONFIG_SCSI_AACRAID=m # CONFIG_MEGARAID_LEGACY is not set CONFIG_MEGARAID_SAS=m CONFIG_SCSI_MPT3SAS=m +CONFIG_SCSI_3SNIC_SSSRAID=m CONFIG_SCSI_MPT2SAS_MAX_SGE=128 CONFIG_SCSI_MPT3SAS_MAX_SGE=128 CONFIG_SCSI_MPT2SAS=m diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig index 0fbe4edeccd0..7339d6179e5a 100644 --- a/drivers/scsi/Kconfig +++ b/drivers/scsi/Kconfig @@ -481,6 +481,7 @@ config SCSI_ARCMSR source "drivers/scsi/esas2r/Kconfig" source "drivers/scsi/megaraid/Kconfig.megaraid" +source "drivers/scsi/sssraid/Kconfig" source "drivers/scsi/mpt3sas/Kconfig" source "drivers/scsi/smartpqi/Kconfig" source "drivers/scsi/ufs/Kconfig" diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile index 78a3c832394c..01bb908fb832 100644 --- a/drivers/scsi/Makefile +++ b/drivers/scsi/Makefile @@ -92,6 +92,7 @@ obj-$(CONFIG_SCSI_CHELSIO_FCOE) += csiostor/ obj-$(CONFIG_SCSI_DMX3191D) += dmx3191d.o obj-$(CONFIG_SCSI_HPSA) += hpsa.o obj-$(CONFIG_SCSI_SMARTPQI) += smartpqi/ +obj-$(CONFIG_SCSI_3SNIC_SSSRAID) += sssraid/ obj-$(CONFIG_SCSI_SYM53C8XX_2) += sym53c8xx_2/ obj-$(CONFIG_SCSI_ZALON) += zalon7xx.o obj-$(CONFIG_SCSI_DC395x) += dc395x.o diff --git a/drivers/scsi/sssraid/Kconfig b/drivers/scsi/sssraid/Kconfig new file mode 100644 index 000000000000..769f68732bf3 --- /dev/null +++ b/drivers/scsi/sssraid/Kconfig @@ -0,0 +1,14 @@ +# +# Kernel configuration file for the 3SNIC +# + +config SCSI_3SNIC_SSSRAID + tristate "3SNIC sssraid Adapter" + depends on PCI && SCSI + select BLK_DEV_BSGLIB + depends on ARM64 || X86_64 + help + This driver supports 3SNIC 3S5xx serial RAID controller, which has + PCI Express Gen4 interface with host and supports SAS/SATA HDD/SSD. + To compile this driver as a module, choose M here: the module will + be called sssraid. diff --git a/drivers/scsi/sssraid/Makefile b/drivers/scsi/sssraid/Makefile new file mode 100644 index 000000000000..c80605866b42 --- /dev/null +++ b/drivers/scsi/sssraid/Makefile @@ -0,0 +1,7 @@ +# +# Makefile for the 3SNIC sssraid drivers. +# + +obj-$(CONFIG_SCSI_3SNIC_SSSRAID) += sssraid.o + +sssraid-objs := sssraid_os.o sssraid_fw.o diff --git a/drivers/scsi/sssraid/sssraid.h b/drivers/scsi/sssraid/sssraid.h new file mode 100644 index 000000000000..bae668ef9c98 --- /dev/null +++ b/drivers/scsi/sssraid/sssraid.h @@ -0,0 +1,971 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2022 3SNIC Information Technology, Ltd */ + +/* 3SNIC RAID SSSXXX Series Linux Driver */ + +#ifndef __SSSRAID_H_ +#define __SSSRAID_H_ + +#define SSSRAID_DRIVER_VERSION "1.0.0.0" +#define SSSRAID_DRIVER_RELDATE "12-December-2022" + +#define SSSRAID_DRIVER_NAME "sssraid" + +#define SSSRAID_NAME_LENGTH 32 + +/* + * SSSRAID Vendor ID and Device IDs + */ +#define PCI_VENDOR_ID_3SNIC_LOGIC 0x1F3F + +#define SSSRAID_SERVER_DEVICE_HBA_DID 0x2100 +#define SSSRAID_SERVER_DEVICE_RAID_DID 0x2200 + +#define SSSRAID_CAP_MQES(cap) ((cap) & 0xffff) +#define SSSRAID_CAP_STRIDE(cap) (((cap) >> 32) & 0xf) +#define SSSRAID_CAP_MPSMIN(cap) (((cap) >> 48) & 0xf) +#define SSSRAID_CAP_MPSMAX(cap) (((cap) >> 52) & 0xf) +#define SSSRAID_CAP_TIMEOUT(cap) (((cap) >> 24) & 0xff) +#define SSSRAID_CAP_DMAMASK(cap) (((cap) >> 37) & 0xff) + +#define SSSRAID_DEFAULT_MAX_CHANNEL 4 +#define SSSRAID_DEFAULT_MAX_ID 240 +#define SSSRAID_DEFAULT_MAX_LUN_PER_HOST 8 +#define MAX_SECTORS 2048 + +/* + * Time define + */ +#define SSSRAID_WATCHDOG_INTERVAL 1000 /* in milli seconds */ +#define SSSRAID_PORTENABLE_TIMEOUT 300 + +/* + * SSSRAID queue and entry size for Admin and I/O type + */ +#define IO_SQE_SIZE sizeof(struct sssraid_ioq_command) +#define ADMIN_SQE_SIZE sizeof(struct sssraid_admin_command) +#define SQE_SIZE(qid) (((qid) > 0) ? IO_SQE_SIZE : ADMIN_SQE_SIZE) +#define CQ_SIZE(depth) ((depth) * sizeof(struct sssraid_completion)) +#define SQ_SIZE(qid, depth) ((depth) * SQE_SIZE(qid)) + +#define SENSE_SIZE(depth) ((depth) * SCSI_SENSE_BUFFERSIZE) + +#define SSSRAID_AQ_DEPTH 128 +#define SSSRAID_NR_AEN_COMMANDS 16 +#define SSSRAID_AQ_BLK_MQ_DEPTH (SSSRAID_AQ_DEPTH - SSSRAID_NR_AEN_COMMANDS) +#define SSSRAID_AQ_MQ_TAG_DEPTH (SSSRAID_AQ_BLK_MQ_DEPTH - 1) + +#define SSSRAID_ADMIN_QUEUE_NUM 1 +#define SSSRAID_PTCMDS_PERQ 1 +#define SSSRAID_IO_BLK_MQ_DEPTH (hdev->scsi_qd) +#define SSSRAID_NR_IOQ_PTCMDS (SSSRAID_PTCMDS_PERQ * hdev->shost->nr_hw_queues) + +#define FUA_MASK 0x08 +#define SSSRAID_MINORS BIT(MINORBITS) +#define SSSRAID_RW_FUA BIT(14) + +#define COMMAND_IS_WRITE(cmd) ((cmd)->common.opcode & 1) + +#define SSSRAID_IO_IOSQES 7 +#define SSSRAID_IO_IOCQES 4 +#define PRP_ENTRY_SIZE 8 + +#define SMALL_POOL_SIZE 256 +#define MAX_SMALL_POOL_NUM 16 +#define MAX_CMD_PER_DEV 64 +#define MAX_CDB_LEN 16 + +#define SSSRAID_UP_TO_MULTY4(x) (((x) + 4) & (~0x03)) + +#define CQE_STATUS_SUCCESS (0x0) + +#define IO_6_DEFAULT_TX_LEN 256 + +#define SGES_PER_PAGE (PAGE_SIZE / sizeof(struct sssraid_sgl_desc)) + +#define SSSRAID_CAP_TIMEOUT_UNIT_MS (HZ / 2) + +extern u32 admin_tmout; +#define ADMIN_TIMEOUT (admin_tmout * HZ) + +#define SSSRAID_WAIT_ABNL_CMD_TIMEOUT 6 + +#define SSSRAID_DMA_MSK_BIT_MAX 64 + +enum { + SSSRAID_SGL_FMT_DATA_DESC = 0x00, + SSSRAID_SGL_FMT_SEG_DESC = 0x02, + SSSRAID_SGL_FMT_LAST_SEG_DESC = 0x03, + SSSRAID_KEY_SGL_FMT_DATA_DESC = 0x04, + SSSRAID_TRANSPORT_SGL_DATA_DESC = 0x05 +}; + + +enum { + SSSRAID_REQ_CANCELLED = (1 << 0), + SSSRAID_REQ_USERCMD = (1 << 1), +}; + +enum { + SSSRAID_SC_SUCCESS = 0x0, + SSSRAID_SC_INVALID_OPCODE = 0x1, + SSSRAID_SC_INVALID_FIELD = 0x2, + + SSSRAID_SC_ABORT_LIMIT = 0x103, + SSSRAID_SC_ABORT_MISSING = 0x104, + SSSRAID_SC_ASYNC_LIMIT = 0x105, + + SSSRAID_SC_DNR = 0x4000, +}; + +enum { + SSSRAID_REG_CAP = 0x0000, + SSSRAID_REG_CC = 0x0014, + SSSRAID_REG_CSTS = 0x001c, + SSSRAID_REG_AQA = 0x0024, + SSSRAID_REG_ASQ = 0x0028, + SSSRAID_REG_ACQ = 0x0030, + SSSRAID_REG_DBS = 0x1000, +}; + +enum { + SSSRAID_CC_ENABLE = 1 << 0, + SSSRAID_CC_CSS_NVM = 0 << 4, + SSSRAID_CC_MPS_SHIFT = 7, + SSSRAID_CC_AMS_SHIFT = 11, + SSSRAID_CC_SHN_SHIFT = 14, + SSSRAID_CC_IOSQES_SHIFT = 16, + SSSRAID_CC_IOCQES_SHIFT = 20, + SSSRAID_CC_AMS_RR = 0 << SSSRAID_CC_AMS_SHIFT, + SSSRAID_CC_SHN_NONE = 0 << SSSRAID_CC_SHN_SHIFT, + SSSRAID_CC_IOSQES = SSSRAID_IO_IOSQES << SSSRAID_CC_IOSQES_SHIFT, + SSSRAID_CC_IOCQES = SSSRAID_IO_IOCQES << SSSRAID_CC_IOCQES_SHIFT, + SSSRAID_CC_SHN_NORMAL = 1 << SSSRAID_CC_SHN_SHIFT, + SSSRAID_CC_SHN_MASK = 3 << SSSRAID_CC_SHN_SHIFT, + SSSRAID_CSTS_CFS_SHIFT = 1, + SSSRAID_CSTS_SHST_SHIFT = 2, + SSSRAID_CSTS_PP_SHIFT = 5, + SSSRAID_CSTS_RDY = 1 << 0, + SSSRAID_CSTS_SHST_CMPLT = 2 << 2, + SSSRAID_CSTS_SHST_MASK = 3 << 2, + SSSRAID_CSTS_CFS_MASK = 1 << SSSRAID_CSTS_CFS_SHIFT, + SSSRAID_CSTS_PP_MASK = 1 << SSSRAID_CSTS_PP_SHIFT, +}; + +enum { + SSSRAID_ADMIN_DELETE_SQ = 0x00, + SSSRAID_ADMIN_CREATE_SQ = 0x01, + SSSRAID_ADMIN_DELETE_CQ = 0x04, + SSSRAID_ADMIN_CREATE_CQ = 0x05, + SSSRAID_ADMIN_ABORT_CMD = 0x08, + SSSRAID_ADMIN_SET_FEATURES = 0x09, + SSSRAID_ADMIN_ASYNC_EVENT = 0x0c, + SSSRAID_ADMIN_GET_INFO = 0xc6, + SSSRAID_ADMIN_RESET = 0xc8, +}; + +enum { + SSSRAID_GET_INFO_CTRL = 0, + SSSRAID_GET_INFO_DEV_LIST = 1, +}; + +enum sssraid_scsi_rst_type { + SSSRAID_RESET_TARGET = 0, + SSSRAID_RESET_BUS = 1, +}; + +enum { + SSSRAID_AEN_ERROR = 0, + SSSRAID_AEN_NOTICE = 2, + SSSRAID_AEN_VS = 7, +}; + +enum { + SSSRAID_AEN_DEV_CHANGED = 0x00, + SSSRAID_AEN_FW_ACT_START = 0x01, + SSSRAID_AEN_HOST_PROBING = 0x10, +}; + +enum { + SSSRAID_AEN_TIMESYN = 0x00, + SSSRAID_AEN_FW_ACT_FINISH = 0x02, + SSSRAID_AEN_EVENT_MIN = 0x80, + SSSRAID_AEN_EVENT_MAX = 0xff, +}; + +enum { + SSSRAID_CMD_WRITE = 0x01, + SSSRAID_CMD_READ = 0x02, + + SSSRAID_CMD_NONIO_NONE = 0x80, + SSSRAID_CMD_NONIO_TODEV = 0x81, + SSSRAID_CMD_NONIO_FROMDEV = 0x82, +}; + +enum { + SSSRAID_QUEUE_PHYS_CONTIG = (1 << 0), + SSSRAID_CQ_IRQ_ENABLED = (1 << 1), + + SSSRAID_FEAT_NUM_QUEUES = 0x07, + SSSRAID_FEAT_ASYNC_EVENT = 0x0b, + SSSRAID_FEAT_TIMESTAMP = 0x0e, +}; + +enum sssraid_state { + SSSRAID_NEW, + SSSRAID_LIVE, + SSSRAID_RESETTING, + SSSRAID_DELETING, + SSSRAID_DEAD, +}; + +enum { + SSSRAID_CARD_HBA, + SSSRAID_CARD_RAID, +}; + +enum sssraid_cmd_type { + SSSRAID_CMD_ADM, + SSSRAID_CMD_IOPT, +}; + +/* + * SSSRAID completion queue entry struct + */ +struct sssraid_completion { + __le32 result; + union { + struct { + __u8 sense_len; + __u8 resv[3]; + }; + __le32 result1; + }; + __le16 sq_head; + __le16 sq_id; + __le16 cmd_id; + __le16 status; +}; + +/* + * SSSRAID firmware controller properties + */ +struct sssraid_ctrl_info { + __le32 nd; + __le16 max_cmds; + __le16 max_channel; + __le32 max_tgt_id; + __le16 max_lun; + __le16 max_num_sge; + __le16 lun_num_in_boot; + __u8 mdts; + __u8 acl; + __u8 aerl; + __u8 card_type; + __u16 rsvd; + __le32 rtd3e; + __u8 sn[32]; + __u8 fr[16]; + __u8 rsvd1[4020]; +}; + +struct sssraid_intr_info { + struct sssraid_dev *hdev; + u16 msix_index; + struct sssraid_cqueue *cqinfo; + char name[SSSRAID_NAME_LENGTH]; +}; + +struct sssraid_fwevt { + struct list_head list; + struct work_struct work; + struct sssraid_dev *hdev; + u16 event_id; + bool send_ack; + bool process_evt; + u32 evt_ctx; + struct kref ref_count; + char event_data[0] __aligned(4); +}; + +/* + * SSSRAID private device struct definition + */ +struct sssraid_dev { + struct pci_dev *pdev; + struct Scsi_Host *shost; + struct sssraid_squeue *sqinfo; + struct sssraid_cqueue *cqinfo; + struct dma_pool *prp_page_pool; + struct dma_pool *prp_small_pool[MAX_SMALL_POOL_NUM]; + void __iomem *bar; + + u32 init_done_queue_cnt; + u32 ioq_depth; + u32 db_stride; + u32 __iomem *dbs; + struct rw_semaphore devices_rwsem; + int numa_node; + u32 page_size; + u32 ctrl_config; + u64 cap; + u32 instance; + u32 scsi_qd; + struct sssraid_ctrl_info *ctrl_info; + struct sssraid_dev_info *devices; + + int logging_level; + + char name[SSSRAID_NAME_LENGTH]; + int cpu_count; + /* + * before_affinity_msix_cnt is + * min("FW support IO Queue count", num_online_cpus)+1 + */ + u16 before_affinity_msix_cnt; + + struct sssraid_cmd *adm_cmds; + struct list_head adm_cmd_list; + spinlock_t adm_cmd_lock; + + struct sssraid_cmd *ioq_ptcmds; + struct list_head ioq_pt_list; + spinlock_t ioq_pt_lock; + + int reset_flag; + + enum sssraid_state state; + spinlock_t state_lock; + + struct request_queue *bsg_queue; + + u8 intr_enabled; + + struct sssraid_intr_info *intr_info; + u32 intr_info_count; + + char watchdog_work_q_name[20]; + struct workqueue_struct *watchdog_work_q; + struct delayed_work watchdog_work; + spinlock_t watchdog_lock; + + char fwevt_worker_name[SSSRAID_NAME_LENGTH]; + struct workqueue_struct *fwevt_worker_thread; + spinlock_t fwevt_lock; + struct list_head fwevt_list; + + struct sssraid_fwevt *current_event; +}; + +/* + * SSSRAID scatter list descriptor + */ +struct sssraid_sgl_desc { + __le64 addr; + __le32 length; + __u8 rsvd[3]; + __u8 type; +}; + +union sssraid_data_ptr { + struct { + __le64 prp1; + __le64 prp2; + }; + struct sssraid_sgl_desc sgl; +}; + +/* + * SSSRAID general admin class command format struct + */ +struct sssraid_admin_common_command { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __le32 cdw2[4]; + union sssraid_data_ptr dptr; + __le32 cdw10; + __le32 cdw11; + __le32 cdw12; + __le32 cdw13; + __le32 cdw14; + __le32 cdw15; +}; + +struct sssraid_features { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __u64 rsvd2[2]; + union sssraid_data_ptr dptr; + __le32 fid; + __le32 dword11; + __le32 dword12; + __le32 dword13; + __le32 dword14; + __le32 dword15; +}; + +/* + * SSSRAID create completion queue command struct + */ +struct sssraid_create_cq { + __u8 opcode; + __u8 flags; + __le16 command_id; + __u32 rsvd1[5]; + __le64 prp1; + __u64 rsvd8; + __le16 cqid; + __le16 qsize; + __le16 cq_flags; + __le16 irq_vector; + __u32 rsvd12[4]; +}; + +/* + * SSSRAID create submission queue command struct + */ +struct sssraid_create_sq { + __u8 opcode; + __u8 flags; + __le16 command_id; + __u32 rsvd1[5]; + __le64 prp1; + __u64 rsvd8; + __le16 sqid; + __le16 qsize; + __le16 sq_flags; + __le16 cqid; + __u32 rsvd12[4]; +}; + +/* + * SSSRAID delete submission queue command struct + */ +struct sssraid_delete_queue { + __u8 opcode; + __u8 flags; + __le16 command_id; + __u32 rsvd1[9]; + __le16 qid; + __u16 rsvd10; + __u32 rsvd11[5]; +}; + +/* + * SSSRAID access to information command struct + */ +struct sssraid_get_info { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __u32 rsvd2[4]; + union sssraid_data_ptr dptr; + __u8 type; + __u8 rsvd10[3]; + __le32 cdw11; + __u32 rsvd12[4]; +}; + +/* + * User command struct + */ +struct sssraid_usr_cmd { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + union { + struct { + __le16 subopcode; + __le16 rsvd1; + } info_0; + __le32 cdw2; + }; + union { + struct { + __le16 data_len; + __le16 param_len; + } info_1; + __le32 cdw3; + }; + __u64 metadata; + union sssraid_data_ptr dptr; + __le32 cdw10; + __le32 cdw11; + __le32 cdw12; + __le32 cdw13; + __le32 cdw14; + __le32 cdw15; +}; + +enum { + SSSRAID_CMD_FLAG_SGL_METABUF = (1 << 6), + SSSRAID_CMD_FLAG_SGL_METASEG = (1 << 7), + SSSRAID_CMD_FLAG_SGL_ALL = SSSRAID_CMD_FLAG_SGL_METABUF | SSSRAID_CMD_FLAG_SGL_METASEG, +}; + +enum sssraid_cmd_state { + SSSRAID_CMD_IDLE = 0, + SSSRAID_CMD_IN_FLIGHT = 1, + SSSRAID_CMD_COMPLETE = 2, + SSSRAID_CMD_TIMEOUT = 3, + SSSRAID_CMD_TMO_COMPLETE = 4, +}; + +/* + * SSSRAID abort command struct + */ +struct sssraid_abort_cmd { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __u64 rsvd2[4]; + __le16 sqid; + __le16 cid; + __u32 rsvd11[5]; +}; + +/* + * SSSRAID reset command struct + */ +struct sssraid_reset_cmd { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __u64 rsvd2[4]; + __u8 type; + __u8 rsvd10[3]; + __u32 rsvd11[5]; +}; + +/* + * SSSRAID admin class command set struct + */ +struct sssraid_admin_command { + union { + struct sssraid_admin_common_command common; + struct sssraid_features features; + struct sssraid_create_cq create_cq; + struct sssraid_create_sq create_sq; + struct sssraid_delete_queue delete_queue; + struct sssraid_get_info get_info; + struct sssraid_abort_cmd abort; + struct sssraid_reset_cmd reset; + struct sssraid_usr_cmd usr_cmd; + }; +}; + +/* + * SSSRAID general IO class command format struct + */ +struct sssraid_ioq_common_command { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __le16 sense_len; + __u8 cdb_len; + __u8 rsvd2; + __le32 cdw3[3]; + union sssraid_data_ptr dptr; + __le32 cdw10[6]; + __u8 cdb[32]; + __le64 sense_addr; + __le32 cdw26[6]; +}; + +/* + * SSSRAID read or write command struct + */ +struct sssraid_rw_command { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __le16 sense_len; + __u8 cdb_len; + __u8 rsvd2; + __u32 rsvd3[3]; + union sssraid_data_ptr dptr; + __le64 slba; + __le16 nlb; + __le16 control; + __u32 rsvd13[3]; + __u8 cdb[32]; + __le64 sense_addr; + __u32 rsvd26[6]; +}; + +struct sssraid_scsi_nonio { + __u8 opcode; + __u8 flags; + __le16 command_id; + __le32 hdid; + __le16 sense_len; + __u8 cdb_length; + __u8 rsvd2; + __u32 rsvd3[3]; + union sssraid_data_ptr dptr; + __u32 rsvd10[5]; + __le32 buffer_len; + __u8 cdb[32]; + __le64 sense_addr; + __u32 rsvd26[6]; +}; + +/* + * SSSRAID IO class command struct + */ +struct sssraid_ioq_command { + union { + struct sssraid_ioq_common_command common; + struct sssraid_rw_command rw; + struct sssraid_scsi_nonio scsi_nonio; + }; +}; + +/* + * SSSRAID passthru command struct + */ +struct sssraid_passthru_common_cmd { + __u8 opcode; + __u8 flags; + __u16 rsvd0; + __u32 nsid; + union { + struct { + __u16 subopcode; + __u16 rsvd1; + } info_0; + __u32 cdw2; + }; + union { + struct { + __u16 data_len; + __u16 param_len; + } info_1; + __u32 cdw3; + }; + __u64 metadata; + + __u64 addr; + __u64 prp2; + + __u32 cdw10; + __u32 cdw11; + __u32 cdw12; + __u32 cdw13; + __u32 cdw14; + __u32 cdw15; + __u32 timeout_ms; + __u32 result0; + __u32 result1; +}; + +struct sssraid_ioq_passthru_cmd { + __u8 opcode; + __u8 flags; + __u16 rsvd0; + __u32 nsid; + union { + struct { + __u16 res_sense_len; + __u8 cdb_len; + __u8 rsvd0; + } info_0; + __u32 cdw2; + }; + union { + struct { + __u16 subopcode; + __u16 rsvd1; + } info_1; + __u32 cdw3; + }; + union { + struct { + __u16 rsvd; + __u16 param_len; + } info_2; + __u32 cdw4; + }; + __u32 cdw5; + __u64 addr; + __u64 prp2; + union { + struct { + __u16 eid; + __u16 sid; + } info_3; + __u32 cdw10; + }; + union { + struct { + __u16 did; + __u8 did_flag; + __u8 rsvd2; + } info_4; + __u32 cdw11; + }; + __u32 cdw12; + __u32 cdw13; + __u32 cdw14; + __u32 data_len; + __u32 cdw16; + __u32 cdw17; + __u32 cdw18; + __u32 cdw19; + __u32 cdw20; + __u32 cdw21; + __u32 cdw22; + __u32 cdw23; + __u64 sense_addr; + __u32 cdw26[4]; + __u32 timeout_ms; + __u32 result0; + __u32 result1; +}; + +struct sssraid_bsg_request { + u32 msgcode; + u32 control; + union { + struct sssraid_passthru_common_cmd admcmd; + struct sssraid_ioq_passthru_cmd ioqcmd; + }; +}; + +enum { + SSSRAID_BSG_ADM, + SSSRAID_BSG_IOQ, +}; + +/* + * define the transfer command struct + */ +struct sssraid_cmd { + u16 qid; + u16 cid; + u32 result0; + u32 result1; + u16 status; + void *priv; + enum sssraid_cmd_state state; + struct completion cmd_done; + struct list_head list; +}; + +/* + * define the SSSRAID physical queue struct + */ +struct sssraid_squeue { + struct sssraid_dev *hdev; + spinlock_t sq_lock; /* spinlock for lock handling */ + + void *sq_cmds; + + dma_addr_t sq_dma_addr; + u32 __iomem *q_db; + u8 cq_phase; + u8 sqes; + u16 qidx; + u16 sq_tail; + u16 last_cq_head; + u16 q_depth; + void *sense; + dma_addr_t sense_dma_addr; + struct dma_pool *prp_small_pool; +}; + +struct sssraid_cqueue { + struct sssraid_dev *hdev; + + spinlock_t cq_lock ____cacheline_aligned_in_smp; /* spinlock for lock handling */ + + struct sssraid_completion *cqes; + + dma_addr_t cq_dma_addr; + u8 cq_phase; + u16 cq_head; + u16 last_cq_head; +}; + +/* + * define the SSSRAID IO queue descriptor struct + */ +struct sssraid_iod { + struct sssraid_squeue *sqinfo; + enum sssraid_cmd_state state; + int npages; + u32 nsge; + u32 length; + bool use_sgl; + dma_addr_t first_dma; + void *sense; + dma_addr_t sense_dma; + struct scatterlist *sg; + void *list[0]; +}; + +/* + * define the SSSRAID scsi device attribution and information + */ +#define SSSRAID_DEV_INFO_ATTR_BOOT(attr) ((attr) & 0x01) +#define SSSRAID_DEV_INFO_ATTR_VD(attr) (((attr) & 0x02) == 0x0) +#define SSSRAID_DEV_INFO_ATTR_PT(attr) (((attr) & 0x22) == 0x02) +#define SSSRAID_DEV_INFO_ATTR_RAWDISK(attr) ((attr) & 0x20) +#define SSSRAID_DEV_DISK_TYPE(attr) ((attr) & 0x1e) + +#define SSSRAID_DEV_INFO_FLAG_VALID(flag) ((flag) & 0x01) +#define SSSRAID_DEV_INFO_FLAG_CHANGE(flag) ((flag) & 0x02) + +/* + * define the SSSRAID scsi device identifier + */ +enum { + SSSRAID_SAS_HDD_VD = 0x04, + SSSRAID_SATA_HDD_VD = 0x08, + SSSRAID_SAS_SSD_VD = 0x0c, + SSSRAID_SATA_SSD_VD = 0x10, + SSSRAID_NVME_SSD_VD = 0x14, + SSSRAID_SAS_HDD_PD = 0x06, + SSSRAID_SATA_HDD_PD = 0x0a, + SSSRAID_SAS_SSD_PD = 0x0e, + SSSRAID_SATA_SSD_PD = 0x12, + SSSRAID_NVME_SSD_PD = 0x16, +}; + +/* + * define the SSSRAID scsi device queue depth + */ +#define SSSRAID_HDD_PD_QD 64 +#define SSSRAID_HDD_VD_QD 256 +#define SSSRAID_SSD_PD_QD 64 +#define SSSRAID_SSD_VD_QD 256 + +#define BGTASK_TYPE_REBUILD 4 +#define USR_CMD_READ 0xc2 +#define USR_CMD_RDLEN 0x1000 +#define USR_CMD_VDINFO 0x704 +#define USR_CMD_BGTASK 0x504 +#define VDINFO_PARAM_LEN 0x04 + +/* + * SSSRAID virtual device information struct + */ +struct sssraid_vd_info { + __u8 name[32]; + __le16 id; + __u8 rg_id; + __u8 rg_level; + __u8 sg_num; + __u8 sg_disk_num; + __u8 vd_status; + __u8 vd_type; + __u8 rsvd1[4056]; +}; + +#define MAX_REALTIME_BGTASK_NUM 32 + +struct bgtask_info { + __u8 type; + __u8 progress; + __u8 rate; + __u8 rsvd0; + __le16 vd_id; + __le16 time_left; + __u8 rsvd1[4]; +}; + +struct sssraid_bgtask { + __u8 sw; + __u8 task_num; + __u8 rsvd[6]; + struct bgtask_info bgtask[MAX_REALTIME_BGTASK_NUM]; +}; + +/* + * SSSRAID scsi device information struct + */ +struct sssraid_dev_info { + __le32 hdid; + __le16 target; + __u8 channel; + __u8 lun; + __u8 attr; + __u8 flag; + __le16 max_io_kb; +}; + +#define IOQ_PT_DATA_LEN 4096 +#define MAX_DEV_ENTRY_PER_PAGE_4K 340 +struct sssraid_dev_list { + __le32 dev_num; + __u32 rsvd0[3]; + struct sssraid_dev_info devices[MAX_DEV_ENTRY_PER_PAGE_4K]; +}; + +/* + * SSSRAID scsi device host data struct + */ +struct sssraid_sdev_hostdata { + u32 hdid; + u16 max_io_kb; + u8 attr; + u8 flag; + u8 rg_id; + u8 rsvd[3]; +}; + +extern unsigned char small_pool_num; +extern u32 io_queue_depth; +irqreturn_t sssraid_isr_poll(int irq, void *privdata); +bool sssraid_poll_cq(struct sssraid_dev *hdev, u16 qidx, int cid); +void sssraid_submit_cmd(struct sssraid_squeue *sqinfo, const void *cmd); +int sssraid_get_dev_list(struct sssraid_dev *hdev, struct sssraid_dev_info *devices); +int sssraid_submit_admin_sync_cmd(struct sssraid_dev *hdev, struct sssraid_admin_command *cmd, + u32 *result0, u32 *result1, u32 timeout); +int sssraid_send_abort_cmd(struct sssraid_dev *hdev, u32 hdid, u16 qidx, u16 cid); +int sssraid_send_reset_cmd(struct sssraid_dev *hdev, u8 type, u32 hdid); +void sssraid_adm_timeout(struct sssraid_dev *hdev, struct sssraid_cmd *cmd); +int sssraid_init_ioc(struct sssraid_dev *hdev, u8 re_init); +void sssraid_cleanup_ioc(struct sssraid_dev *hdev, u8 re_init); +int sssraid_soft_reset_handler(struct sssraid_dev *hdev); +void sssraid_free_iod_res(struct sssraid_dev *hdev, struct sssraid_iod *iod); +bool sssraid_change_host_state(struct sssraid_dev *hdev, enum sssraid_state newstate); +int sssraid_configure_timestamp(struct sssraid_dev *hdev); +int sssraid_init_ctrl_info(struct sssraid_dev *hdev); +struct sssraid_cmd *sssraid_get_cmd(struct sssraid_dev *hdev, enum sssraid_cmd_type type); +void sssraid_put_cmd(struct sssraid_dev *hdev, struct sssraid_cmd *cmd, + enum sssraid_cmd_type type); +int sssraid_send_event_ack(struct sssraid_dev *hdev, u8 event, + u32 event_ctx, u16 cid); +struct sssraid_fwevt *sssraid_alloc_fwevt(int len); +void sssraid_fwevt_add_to_list(struct sssraid_dev *hdev, + struct sssraid_fwevt *fwevt); +void sssraid_cleanup_fwevt_list(struct sssraid_dev *hdev); +void sssraid_ioc_enable_intr(struct sssraid_dev *hdev); +void sssraid_ioc_disable_intr(struct sssraid_dev *hdev); +void sssraid_cleanup_resources(struct sssraid_dev *hdev); +void sssraid_complete_cqes(struct sssraid_dev *hdev, u16 qidx, u16 start, u16 end); +int sssraid_io_map_data(struct sssraid_dev *hdev, struct sssraid_iod *iod, + struct scsi_cmnd *scmd, struct sssraid_ioq_command *ioq_cmd); +void sssraid_map_status(struct sssraid_iod *iod, struct scsi_cmnd *scmd, + struct sssraid_completion *cqe); +void sssraid_scan_disk(struct sssraid_dev *hdev); +void sssraid_complete_aen(struct sssraid_dev *hdev, struct sssraid_completion *cqe); +void sssraid_back_all_io(struct sssraid_dev *hdev); + +static inline void **sssraid_iod_list(struct sssraid_iod *iod) +{ + return iod->list; +} + +#end diff --git a/drivers/scsi/sssraid/sssraid_debug.h b/drivers/scsi/sssraid/sssraid_debug.h new file mode 100644 index 000000000000..d48c57e275f7 --- /dev/null +++ b/drivers/scsi/sssraid/sssraid_debug.h @@ -0,0 +1,34 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright(c) 2022 3SNIC Information Technology, Ltd */ + +/* 3SNIC RAID SSSXXX Series Linux Driver */ + +#ifndef SSSRAID_DEBUG_H_INCLUDED +#define SSSRAID_DEBUG_H_INCLUDED + +/* + * debug levels + */ +#define SSSRAID_DEBUG 0x00000001 + +/* + * debug macros + */ + +#define ioc_err(ioc, fmt, ...) \ + pr_err("%s: " fmt, (ioc)->name, ##__VA_ARGS__) +#define ioc_notice(ioc, fmt, ...) \ + pr_notice("%s: " fmt, (ioc)->name, ##__VA_ARGS__) +#define ioc_warn(ioc, fmt, ...) \ + pr_warn("%s: " fmt, (ioc)->name, ##__VA_ARGS__) +#define ioc_info(ioc, fmt, ...) \ + pr_info("%s: " fmt, (ioc)->name, ##__VA_ARGS__) + + +#define dbgprint(IOC, FMT, ...) \ + do { \ + if (unlikely(IOC->logging_level & SSSRAID_DEBUG)) \ + pr_info("%s: " FMT, (IOC)->name, ##__VA_ARGS__); \ + } while (0) + +#endif /* SSSRAID_DEBUG_H_INCLUDED */ diff --git a/drivers/scsi/sssraid/sssraid_fw.c b/drivers/scsi/sssraid/sssraid_fw.c new file mode 100644 index 000000000000..54287cf70558 --- /dev/null +++ b/drivers/scsi/sssraid/sssraid_fw.c @@ -0,0 +1,1724 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2022 3SNIC Information Technology, Ltd */ + +/* 3SNIC RAID SSSXXX Series Linux Driver */ + +#include <linux/version.h> +#include <linux/pci.h> +#include <linux/aer.h> +#include <linux/module.h> +#include <linux/ioport.h> +#include <linux/device.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/cdev.h> +#include <linux/sysfs.h> +#include <linux/gfp.h> +#include <linux/types.h> +#include <linux/ratelimit.h> +#include <linux/debugfs.h> +#include <linux/blkdev.h> +#include <linux/bsg-lib.h> +#include <linux/sort.h> + +#include <scsi/scsi.h> +#include <scsi/scsi_cmnd.h> +#include <scsi/scsi_device.h> +#include <scsi/scsi_host.h> +#include <scsi/scsi_transport.h> +#include <scsi/scsi_dbg.h> + +#include <linux/unaligned/be_byteshift.h> +#include <linux/unaligned/le_byteshift.h> +#include <linux/once.h> +#include <linux/sched/signal.h> +#include <linux/io-64-nonatomic-lo-hi.h> + +#include "sssraid.h" +#include "sssraid_debug.h" + +static int sssraid_wait_ready(struct sssraid_dev *hdev, u64 cap, bool enabled) +{ + unsigned long timeout = + ((SSSRAID_CAP_TIMEOUT(cap) + 1) * SSSRAID_CAP_TIMEOUT_UNIT_MS) + jiffies; + u32 bit = enabled ? SSSRAID_CSTS_RDY : 0; + + while ((readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_RDY) != bit) { + usleep_range(1000, 2000); + if (fatal_signal_pending(current)) + return -EINTR; + + if (time_after(jiffies, timeout)) { + ioc_err(hdev, "Device not ready; aborting %s\n", + enabled ? "initialisation" : "reset"); + return -ENODEV; + } + } + return 0; +} + +static int sssraid_enable_ctrl(struct sssraid_dev *hdev) +{ + u64 cap = hdev->cap; + u32 dev_page_min = SSSRAID_CAP_MPSMIN(cap) + 12; + u32 page_shift = PAGE_SHIFT; + + if (page_shift < dev_page_min) { + ioc_err(hdev, "Minimum device page size[%u], too large for host[%u]\n", + 1U << dev_page_min, 1U << page_shift); + return -ENODEV; + } + + page_shift = min_t(unsigned int, SSSRAID_CAP_MPSMAX(cap) + 12, PAGE_SHIFT); + hdev->page_size = 1U << page_shift; + + hdev->ctrl_config = SSSRAID_CC_CSS_NVM; + hdev->ctrl_config |= (page_shift - 12) << SSSRAID_CC_MPS_SHIFT; + hdev->ctrl_config |= SSSRAID_CC_AMS_RR | SSSRAID_CC_SHN_NONE; + hdev->ctrl_config |= SSSRAID_CC_IOSQES | SSSRAID_CC_IOCQES; + hdev->ctrl_config |= SSSRAID_CC_ENABLE; + writel(hdev->ctrl_config, hdev->bar + SSSRAID_REG_CC); + + return sssraid_wait_ready(hdev, cap, true); +} + +static int sssraid_disable_ctrl(struct sssraid_dev *hdev) +{ + hdev->ctrl_config &= ~SSSRAID_CC_SHN_MASK; + hdev->ctrl_config &= ~SSSRAID_CC_ENABLE; + writel(hdev->ctrl_config, hdev->bar + SSSRAID_REG_CC); + + return sssraid_wait_ready(hdev, hdev->cap, false); +} + +static int sssraid_shutdown_ctrl(struct sssraid_dev *hdev) +{ + unsigned long timeout = le32_to_cpu(hdev->ctrl_info->rtd3e) / 1000000 * HZ + jiffies; + + hdev->ctrl_config &= ~SSSRAID_CC_SHN_MASK; + hdev->ctrl_config |= SSSRAID_CC_SHN_NORMAL; + writel(hdev->ctrl_config, hdev->bar + SSSRAID_REG_CC); + + while ((readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_SHST_MASK) != + SSSRAID_CSTS_SHST_CMPLT) { + msleep(100); + if (fatal_signal_pending(current)) + return -EINTR; + if (time_after(jiffies, timeout)) { + ioc_err(hdev, "Device shutdown incomplete; abort shutdown\n"); + return -ENODEV; + } + } + return 0; +} + +static int sssraid_remap_bar(struct sssraid_dev *hdev, u32 size) +{ + struct pci_dev *pdev = hdev->pdev; + + if (size > pci_resource_len(pdev, 0)) { + ioc_err(hdev, "Input size[%u] exceed bar0 length[%llu]\n", + size, pci_resource_len(pdev, 0)); + return -ENODEV; + } + + if (hdev->bar) + iounmap(hdev->bar); + + hdev->bar = ioremap(pci_resource_start(pdev, 0), size); + if (!hdev->bar) { + ioc_err(hdev, "ioremap for bar0 failed\n"); + return -ENODEV; + } + hdev->dbs = hdev->bar + SSSRAID_REG_DBS; + + return 0; +} + +static int sssraid_create_dma_pools(struct sssraid_dev *hdev) +{ + int i; + char poolname[20] = { 0 }; + + hdev->prp_page_pool = dma_pool_create("prp list page", &hdev->pdev->dev, + PAGE_SIZE, PAGE_SIZE, 0); + + if (!hdev->prp_page_pool) { + ioc_err(hdev, "create prp_page_pool failed\n"); + return -ENOMEM; + } + + for (i = 0; i < small_pool_num; i++) { + sprintf(poolname, "prp_list_256_%d", i); + hdev->prp_small_pool[i] = dma_pool_create(poolname, &hdev->pdev->dev, + SMALL_POOL_SIZE, + SMALL_POOL_SIZE, 0); + + if (!hdev->prp_small_pool[i]) { + ioc_err(hdev, "create prp_small_pool %d failed\n", i); + goto destroy_prp_small_pool; + } + } + + return 0; + +destroy_prp_small_pool: + while (i > 0) + dma_pool_destroy(hdev->prp_small_pool[--i]); + dma_pool_destroy(hdev->prp_page_pool); + + return -ENOMEM; +} + +static void sssraid_destroy_dma_pools(struct sssraid_dev *hdev) +{ + int i; + + for (i = 0; i < small_pool_num; i++) + dma_pool_destroy(hdev->prp_small_pool[i]); + dma_pool_destroy(hdev->prp_page_pool); +} + +static int sssraid_alloc_resources(struct sssraid_dev *hdev) +{ + int retval, nqueue; + + hdev->ctrl_info = kzalloc_node(sizeof(*hdev->ctrl_info), GFP_KERNEL, hdev->numa_node); + if (!hdev->ctrl_info) + return -ENOMEM; + + retval = sssraid_create_dma_pools(hdev); + if (retval) + goto free_ctrl_info; + /* not num_online_cpus */ + nqueue = num_possible_cpus() + 1; + hdev->cqinfo = kcalloc_node(nqueue, sizeof(struct sssraid_cqueue), + GFP_KERNEL, hdev->numa_node); + if (!hdev->cqinfo) { + retval = -ENOMEM; + goto destroy_dma_pools; + } + + hdev->sqinfo = kcalloc_node(nqueue, sizeof(struct sssraid_squeue), + GFP_KERNEL, hdev->numa_node); + if (!hdev->sqinfo) { + retval = -ENOMEM; + goto free_cqueues; + } + + /* sssraid_alloc_admin_cmds moved to sssraid_init_ioc */ + + ioc_info(hdev, "queues num: %d\n", nqueue); + + return 0; + +free_cqueues: + kfree(hdev->cqinfo); +destroy_dma_pools: + sssraid_destroy_dma_pools(hdev); +free_ctrl_info: + kfree(hdev->ctrl_info); + + return retval; +} + +void sssraid_ioc_enable_intr(struct sssraid_dev *hdev) +{ + hdev->intr_enabled = 1; +} + +void sssraid_ioc_disable_intr(struct sssraid_dev *hdev) +{ + u16 i, max_vectors; + + hdev->intr_enabled = 0; + max_vectors = hdev->intr_info_count; + + for (i = 0; i < max_vectors; i++) + synchronize_irq(pci_irq_vector(hdev->pdev, i)); +} + +static int sssraid_setup_resources(struct sssraid_dev *hdev) +{ + struct pci_dev *pdev = hdev->pdev; + int retval = 0; + u64 maskbit = SSSRAID_DMA_MSK_BIT_MAX; + + if (pci_enable_device_mem(pdev)) { + ioc_err(hdev, "pci_enable_device_mem: failed\n"); + retval = -ENODEV; + goto out_failed; + } + + retval = pci_request_mem_regions(pdev, SSSRAID_DRIVER_NAME); + if (retval) { + ioc_err(hdev, "fail to request memory regions\n"); + retval = -ENODEV; + goto out_failed; + } + + /* get cap value at first, so keep + * sssraid_remap_bar(hdev, SSSRAID_REG_DBS + 4096) + * ioremap(pci_resource_start(..)) still in sssraid_remap_bar + */ + retval = sssraid_remap_bar(hdev, SSSRAID_REG_DBS + 4096); + if (retval) { + retval = -ENODEV; + goto out_failed; + } + + pci_set_master(pdev); + + if (readl(hdev->bar + SSSRAID_REG_CSTS) == U32_MAX) { + retval = -ENODEV; + ioc_err(hdev, "Read csts register failed\n"); + goto out_failed; + } + + hdev->cap = lo_hi_readq(hdev->bar + SSSRAID_REG_CAP); + hdev->ioq_depth = min_t(u32, SSSRAID_CAP_MQES(hdev->cap) + 1, io_queue_depth); + hdev->scsi_qd = hdev->ioq_depth - SSSRAID_PTCMDS_PERQ; + hdev->db_stride = 1 << SSSRAID_CAP_STRIDE(hdev->cap); + + maskbit = SSSRAID_CAP_DMAMASK(hdev->cap); + if (maskbit < 32 || maskbit > SSSRAID_DMA_MSK_BIT_MAX) { + ioc_err(hdev, "err, dma mask invalid[%llu], set to default\n", maskbit); + maskbit = SSSRAID_DMA_MSK_BIT_MAX; + } + + if (dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(maskbit))) { + if (dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32))) { + ioc_err(hdev, "set dma mask[32] and coherent failed\n"); + retval = -ENODEV; + goto out_failed; + } + ioc_info(hdev, "set dma mask[32] success\n"); + } else { + ioc_info(hdev, "set dma mask[%llu] success\n", maskbit); + } + + /* pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_ALL_TYPES) moved to setup_isr */ + + pci_set_drvdata(pdev, hdev->shost); + + pci_enable_pcie_error_reporting(pdev); + pci_save_state(pdev); + + sssraid_ioc_disable_intr(hdev); + + return retval; + +out_failed: + sssraid_cleanup_resources(hdev); + return retval; +} + +static int sssraid_alloc_admin_cmds(struct sssraid_dev *hdev) +{ + u16 i; + + INIT_LIST_HEAD(&hdev->adm_cmd_list); + spin_lock_init(&hdev->adm_cmd_lock); + + hdev->adm_cmds = kcalloc_node(SSSRAID_AQ_BLK_MQ_DEPTH, sizeof(struct sssraid_cmd), + GFP_KERNEL, hdev->numa_node); + + if (!hdev->adm_cmds) { + ioc_err(hdev, "Alloc admin cmds failed\n"); + return -ENOMEM; + } + + for (i = 0; i < SSSRAID_AQ_BLK_MQ_DEPTH; i++) { + hdev->adm_cmds[i].qid = 0; + hdev->adm_cmds[i].cid = i; + list_add_tail(&(hdev->adm_cmds[i].list), &hdev->adm_cmd_list); + } + + ioc_info(hdev, "Alloc admin cmds success, num[%d]\n", SSSRAID_AQ_BLK_MQ_DEPTH); + + return 0; +} + +static int sssraid_alloc_qpair(struct sssraid_dev *hdev, u16 qidx, u16 depth) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + int retval = 0; + + if (hdev->init_done_queue_cnt > qidx) { + ioc_info(hdev, "warn: queue[%d] is exist\n", qidx); + return 0; + } + + cqinfo->cqes = dma_alloc_coherent(&hdev->pdev->dev, CQ_SIZE(depth), + &cqinfo->cq_dma_addr, GFP_KERNEL | __GFP_ZERO); + if (!cqinfo->cqes) + return -ENOMEM; + + sqinfo->sq_cmds = dma_alloc_coherent(&hdev->pdev->dev, SQ_SIZE(qidx, depth), + &sqinfo->sq_dma_addr, GFP_KERNEL); + if (!sqinfo->sq_cmds) { + retval = -ENOMEM; + goto free_cqes; + } + + /* alloc sense buffer */ + sqinfo->sense = dma_alloc_coherent(&hdev->pdev->dev, SENSE_SIZE(depth), + &sqinfo->sense_dma_addr, GFP_KERNEL | __GFP_ZERO); + if (!sqinfo->sense) { + retval = -ENOMEM; + goto free_sq_cmds; + } + + spin_lock_init(&sqinfo->sq_lock); + spin_lock_init(&cqinfo->cq_lock); + cqinfo->hdev = hdev; + sqinfo->hdev = hdev; + sqinfo->q_depth = depth; + sqinfo->qidx = qidx; + /* cq_vector replaced by msix_index */ + + /* + * online_queues: completely initialized queue count: sssraid_init_queue + * queue_count: allocated but not completely initialized queue count: sssraid_alloc_queue + * online_queues/queue_count replaced by init_done_queue_cnt. + */ + hdev->init_done_queue_cnt++; + + return 0; + +free_sq_cmds: + dma_free_coherent(&hdev->pdev->dev, SQ_SIZE(qidx, depth), (void *)sqinfo->sq_cmds, + sqinfo->sq_dma_addr); +free_cqes: + dma_free_coherent(&hdev->pdev->dev, CQ_SIZE(depth), (void *)cqinfo->cqes, + cqinfo->cq_dma_addr); + return retval; +} + +static void sssraid_init_queue(struct sssraid_dev *hdev, u16 qidx) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + + memset((void *)cqinfo->cqes, 0, CQ_SIZE(sqinfo->q_depth)); + + sqinfo->sq_tail = 0; + cqinfo->cq_head = 0; + cqinfo->cq_phase = 1; + sqinfo->q_db = &hdev->dbs[qidx * 2 * hdev->db_stride]; + sqinfo->prp_small_pool = hdev->prp_small_pool[qidx % small_pool_num]; +} + +static int sssraid_setup_admin_qpair(struct sssraid_dev *hdev) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[0]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[0]; + u32 aqa; + int retval; + + ioc_info(hdev, "start disable ctrl\n"); + + retval = sssraid_disable_ctrl(hdev); + if (retval) + return retval; + + /* this func don't alloc admin queue */ + + aqa = sqinfo->q_depth - 1; + aqa |= aqa << 16; + writel(aqa, hdev->bar + SSSRAID_REG_AQA); + lo_hi_writeq(sqinfo->sq_dma_addr, hdev->bar + SSSRAID_REG_ASQ); + lo_hi_writeq(cqinfo->cq_dma_addr, hdev->bar + SSSRAID_REG_ACQ); + + ioc_info(hdev, "start enable ctrl\n"); + + retval = sssraid_enable_ctrl(hdev); + if (retval) { + retval = -ENODEV; + return retval; + } + + /* interrupt registry not here */ + /* cq_vector replaced by msix_index */ + + sssraid_init_queue(hdev, 0); + + ioc_info(hdev, "success, queuecount:[%d], pagesize[%d]\n", + hdev->init_done_queue_cnt, hdev->page_size); + + return 0; +} + +static void sssraid_cleanup_isr(struct sssraid_dev *hdev) +{ + u16 i; + + sssraid_ioc_disable_intr(hdev); + + if (!hdev->intr_info) + return; + + for (i = 0; i < hdev->intr_info_count; i++) + free_irq(pci_irq_vector(hdev->pdev, i), + (hdev->intr_info + i)); + + kfree(hdev->intr_info); + hdev->intr_info = NULL; + hdev->intr_info_count = 0; + pci_free_irq_vectors(hdev->pdev); +} + +static void sssraid_complete_adminq_cmnd(struct sssraid_dev *hdev, u16 qidx, + struct sssraid_completion *cqe) +{ + struct sssraid_cmd *adm_cmd; + + adm_cmd = hdev->adm_cmds + le16_to_cpu(cqe->cmd_id); + if (unlikely(adm_cmd->state == SSSRAID_CMD_IDLE)) { + ioc_warn(hdev, "Invalid id %d completed on queue %d\n", + le16_to_cpu(cqe->cmd_id), le16_to_cpu(cqe->sq_id)); + return; + } + + adm_cmd->status = le16_to_cpu(cqe->status) >> 1; + adm_cmd->result0 = le32_to_cpu(cqe->result); + adm_cmd->result1 = le32_to_cpu(cqe->result1); + + complete(&adm_cmd->cmd_done); +} + +static inline bool sssraid_cqe_pending(struct sssraid_cqueue *cqinfo) +{ + return (le16_to_cpu(cqinfo->cqes[cqinfo->cq_head].status) & 1) == + cqinfo->cq_phase; +} + +static inline void sssraid_update_cq_head(struct sssraid_dev *hdev, u16 qidx) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + + if (++cqinfo->cq_head == sqinfo->q_depth) { + cqinfo->cq_head = 0; + cqinfo->cq_phase = !cqinfo->cq_phase; + } +} + +static inline bool sssraid_process_cq(struct sssraid_dev *hdev, u16 qidx, u16 *start, + u16 *end, int tag) +{ + bool found = false; + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + + *start = cqinfo->cq_head; + while (!found && sssraid_cqe_pending(cqinfo)) { + if (le16_to_cpu(cqinfo->cqes[cqinfo->cq_head].cmd_id) == tag) + found = true; + sssraid_update_cq_head(hdev, qidx); + } + *end = cqinfo->cq_head; + + if (*start != *end) + writel(cqinfo->cq_head, sqinfo->q_db + sqinfo->hdev->db_stride); + + return found; +} + +static irqreturn_t sssraid_isr(int irq, void *privdata) +{ + struct sssraid_intr_info *intr_info = privdata; + struct sssraid_dev *hdev = intr_info->hdev; + irqreturn_t ret = IRQ_NONE; + struct sssraid_cqueue *cqinfo; + u16 midx, start, end; + + if (!intr_info) + return IRQ_NONE; + + if (!hdev->intr_enabled) + return IRQ_NONE; + + midx = intr_info->msix_index; + cqinfo = &hdev->cqinfo[midx]; + + spin_lock(&cqinfo->cq_lock); + if (cqinfo->cq_head != cqinfo->last_cq_head) + ret = IRQ_HANDLED; + + sssraid_process_cq(hdev, midx, &start, &end, -1); + cqinfo->last_cq_head = cqinfo->cq_head; + spin_unlock(&cqinfo->cq_lock); + + if (start != end) { + sssraid_complete_cqes(hdev, midx, start, end); + ret = IRQ_HANDLED; + } + return ret; +} + +irqreturn_t sssraid_isr_poll(int irq, void *privdata) +{ + return IRQ_NONE; +} + +bool sssraid_poll_cq(struct sssraid_dev *hdev, u16 qidx, int cid) +{ + u16 start, end; + bool found; + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + + if (!sssraid_cqe_pending(cqinfo)) + return 0; + + spin_lock_irq(&cqinfo->cq_lock); + found = sssraid_process_cq(hdev, qidx, &start, &end, cid); + spin_unlock_irq(&cqinfo->cq_lock); + + sssraid_complete_cqes(hdev, qidx, start, end); + return found; +} + +static inline int sssraid_request_irq(struct sssraid_dev *hdev, u16 index) +{ + struct pci_dev *pdev = hdev->pdev; + struct sssraid_intr_info *intr_info = hdev->intr_info + index; + int retval = 0; + + intr_info->hdev = hdev; + intr_info->msix_index = index; + intr_info->cqinfo = NULL; + + snprintf(intr_info->name, SSSRAID_NAME_LENGTH, "%s%d-msix%d", + SSSRAID_DRIVER_NAME, hdev->instance, index); + + retval = request_threaded_irq(pci_irq_vector(pdev, index), sssraid_isr, + sssraid_isr_poll, IRQF_SHARED, intr_info->name, intr_info); + + if (retval) { + ioc_err(hdev, "%s: Unable to allocate interrupt %d!\n", + intr_info->name, pci_irq_vector(pdev, index)); + return retval; + } + + return retval; +} + +static int sssraid_setup_isr(struct sssraid_dev *hdev, u8 setup_one) +{ + unsigned int irq_flags = PCI_IRQ_MSIX; + u16 max_vectors = 0, i; + int retval = 0; + + struct irq_affinity desc = { .pre_vectors = 1}; + + sssraid_cleanup_isr(hdev); + + if (setup_one) + max_vectors = 1; + else { + max_vectors = hdev->before_affinity_msix_cnt; + + ioc_info(hdev, "MSI-x vectors requested: %d\n", max_vectors); + } + + irq_flags |= PCI_IRQ_AFFINITY | PCI_IRQ_ALL_TYPES; + + i = pci_alloc_irq_vectors_affinity(hdev->pdev, + 1, max_vectors, irq_flags, &desc); + + if (i <= 0) { + ioc_err(hdev, "Cannot alloc irq vectors\n"); + goto out_failed; + } + if (i != max_vectors) { + ioc_info(hdev, + "allocated vectors (%d) are less than configured (%d)\n", + i, max_vectors); + + max_vectors = i; + } + + hdev->intr_info = kzalloc(sizeof(struct sssraid_intr_info) * max_vectors, + GFP_KERNEL); + if (!hdev->intr_info) { + retval = -1; + pci_free_irq_vectors(hdev->pdev); + goto out_failed; + } + + for (i = 0; i < max_vectors; i++) { + retval = sssraid_request_irq(hdev, i); + if (retval) { + hdev->intr_info_count = i; /* =i is for offload interrupt loop counter */ + goto out_failed; + } + } + + /* intr_info_count replace max_qid */ + hdev->intr_info_count = max_vectors; + sssraid_ioc_enable_intr(hdev); + return retval; +out_failed: + sssraid_cleanup_isr(hdev); + + return retval; +} + +static bool sssraid_adm_need_reset(struct sssraid_admin_command *cmd) +{ + switch (cmd->common.opcode) { + case SSSRAID_ADMIN_DELETE_SQ: + case SSSRAID_ADMIN_CREATE_SQ: + case SSSRAID_ADMIN_DELETE_CQ: + case SSSRAID_ADMIN_CREATE_CQ: + case SSSRAID_ADMIN_SET_FEATURES: + return false; + default: + return true; + } +} + +void sssraid_submit_cmd(struct sssraid_squeue *sqinfo, const void *cmd) +{ + u32 sqes = SQE_SIZE(sqinfo->qidx); + unsigned long flags; + struct sssraid_admin_common_command *acd = (struct sssraid_admin_common_command *)cmd; + + spin_lock_irqsave(&sqinfo->sq_lock, flags); + memcpy((sqinfo->sq_cmds + sqes * sqinfo->sq_tail), cmd, sqes); + if (++sqinfo->sq_tail == sqinfo->q_depth) + sqinfo->sq_tail = 0; + + writel(sqinfo->sq_tail, sqinfo->q_db); + spin_unlock_irqrestore(&sqinfo->sq_lock, flags); + + dbgprint(sqinfo->hdev, "cid[%d] qidx[%d], opcode[0x%x], flags[0x%x], hdid[%u]\n", + le16_to_cpu(acd->command_id), sqinfo->qidx, acd->opcode, acd->flags, + le32_to_cpu(acd->hdid)); +} + +int sssraid_submit_admin_sync_cmd(struct sssraid_dev *hdev, struct sssraid_admin_command *cmd, + u32 *result0, u32 *result1, u32 timeout) +{ + struct sssraid_cmd *adm_cmd = sssraid_get_cmd(hdev, SSSRAID_CMD_ADM); + + if (!adm_cmd) { + ioc_err(hdev, "err, get admin cmd failed\n"); + return -EFAULT; + } + + timeout = timeout ? timeout : ADMIN_TIMEOUT; + + /* + * watch dog not as optimized as + * init_completion/complete + */ + init_completion(&adm_cmd->cmd_done); + + cmd->common.command_id = cpu_to_le16(adm_cmd->cid); + sssraid_submit_cmd(&hdev->sqinfo[0], cmd); + + if (!wait_for_completion_timeout(&adm_cmd->cmd_done, timeout)) { + ioc_err(hdev, "cid[%d] qidx[%d] timeout, opcode[0x%x] subopcode[0x%x]\n", + adm_cmd->cid, adm_cmd->qid, cmd->usr_cmd.opcode, + cmd->usr_cmd.info_0.subopcode); + + /* reset controller if admin timeout */ + if (sssraid_adm_need_reset(cmd)) + sssraid_adm_timeout(hdev, adm_cmd); + + sssraid_put_cmd(hdev, adm_cmd, SSSRAID_CMD_ADM); + return -ETIME; + } + + if (result0) + *result0 = adm_cmd->result0; + if (result1) + *result1 = adm_cmd->result1; + + sssraid_put_cmd(hdev, adm_cmd, SSSRAID_CMD_ADM); + + return adm_cmd->status; +} + +static int sssraid_get_ctrl_info(struct sssraid_dev *hdev, struct sssraid_ctrl_info *ctrl_info) +{ + struct sssraid_admin_command admin_cmd; + u8 *data_ptr = NULL; + dma_addr_t data_dma = 0; + int retval; + + data_ptr = dma_alloc_coherent(&hdev->pdev->dev, PAGE_SIZE, &data_dma, GFP_KERNEL); + if (!data_ptr) + return -ENOMEM; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.get_info.opcode = SSSRAID_ADMIN_GET_INFO; + admin_cmd.get_info.type = SSSRAID_GET_INFO_CTRL; + admin_cmd.common.dptr.prp1 = cpu_to_le64(data_dma); + + retval = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); + if (!retval) + memcpy(ctrl_info, data_ptr, sizeof(struct sssraid_ctrl_info)); + + dma_free_coherent(&hdev->pdev->dev, PAGE_SIZE, data_ptr, data_dma); + + return retval; +} + +int sssraid_init_ctrl_info(struct sssraid_dev *hdev) +{ + int retval; + + hdev->ctrl_info->nd = cpu_to_le32(240); + hdev->ctrl_info->mdts = 8; + hdev->ctrl_info->max_cmds = cpu_to_le16(4096); + hdev->ctrl_info->max_num_sge = cpu_to_le16(128); + hdev->ctrl_info->max_channel = cpu_to_le16(4); + hdev->ctrl_info->max_tgt_id = cpu_to_le32(3239); + hdev->ctrl_info->max_lun = cpu_to_le16(2); + + retval = sssraid_get_ctrl_info(hdev, hdev->ctrl_info); + if (retval) + ioc_err(hdev, "init controller info failed: %d\n", retval); + + ioc_info(hdev, "nd = %d\n", le32_to_cpu(hdev->ctrl_info->nd)); + ioc_info(hdev, "max_cmd = %d\n", le16_to_cpu(hdev->ctrl_info->max_cmds)); + ioc_info(hdev, "max_channel = %d\n", le16_to_cpu(hdev->ctrl_info->max_channel)); + ioc_info(hdev, "max_tgt_id = %d\n", le32_to_cpu(hdev->ctrl_info->max_tgt_id)); + ioc_info(hdev, "max_lun = %d\n", le16_to_cpu(hdev->ctrl_info->max_lun)); + ioc_info(hdev, "max_num_sge = %d\n", le16_to_cpu(hdev->ctrl_info->max_num_sge)); + ioc_info(hdev, "lun_num_boot = %d\n", le16_to_cpu(hdev->ctrl_info->lun_num_in_boot)); + ioc_info(hdev, "mdts = %d\n", hdev->ctrl_info->mdts); + ioc_info(hdev, "acl = %d\n", hdev->ctrl_info->acl); + ioc_info(hdev, "aer1 = %d\n", hdev->ctrl_info->aerl); + ioc_info(hdev, "card_type = %d\n", hdev->ctrl_info->card_type); + ioc_info(hdev, "rtd3e = %d\n", le32_to_cpu(hdev->ctrl_info->rtd3e)); + ioc_info(hdev, "sn = %s\n", hdev->ctrl_info->sn); + ioc_info(hdev, "fr = %s\n", hdev->ctrl_info->fr); + + if (!hdev->ctrl_info->aerl) + hdev->ctrl_info->aerl = 1; + if (hdev->ctrl_info->aerl > SSSRAID_NR_AEN_COMMANDS) + hdev->ctrl_info->aerl = SSSRAID_NR_AEN_COMMANDS; + + return 0; +} + +static int sssraid_set_features(struct sssraid_dev *hdev, u32 fid, u32 dword11, void *buffer, + size_t buflen, u32 *result) +{ + struct sssraid_admin_command admin_cmd; + int ret; + u8 *data_ptr = NULL; + dma_addr_t data_dma = 0; + + if (buffer && buflen) { + data_ptr = dma_alloc_coherent(&hdev->pdev->dev, buflen, &data_dma, GFP_KERNEL); + if (!data_ptr) + return -ENOMEM; + + memcpy(data_ptr, buffer, buflen); + } + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.features.opcode = SSSRAID_ADMIN_SET_FEATURES; + admin_cmd.features.fid = cpu_to_le32(fid); + admin_cmd.features.dword11 = cpu_to_le32(dword11); + admin_cmd.common.dptr.prp1 = cpu_to_le64(data_dma); + + ret = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, result, NULL, 0); + + if (data_ptr) + dma_free_coherent(&hdev->pdev->dev, buflen, data_ptr, data_dma); + + return ret; +} + +static int sssraid_set_queue_cnt(struct sssraid_dev *hdev, u32 *cnt) +{ + u32 q_cnt = (*cnt - 1) | ((*cnt - 1) << 16); + u32 nr_ioqs, result; + int status; + + status = sssraid_set_features(hdev, SSSRAID_FEAT_NUM_QUEUES, q_cnt, NULL, 0, &result); + if (status) { + ioc_err(hdev, "Set queue count failed, status: %d\n", + status); + return -EIO; + } + + nr_ioqs = min(result & 0xffff, result >> 16) + 1; + *cnt = min(*cnt, nr_ioqs); + if (*cnt == 0) { + ioc_err(hdev, "Illegal queue count: zero\n"); + return -EIO; + } + return 0; +} + +static int sssraid_create_cq(struct sssraid_dev *hdev, u16 qidx) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[qidx]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + struct sssraid_admin_command admin_cmd; + int flags = SSSRAID_QUEUE_PHYS_CONTIG | SSSRAID_CQ_IRQ_ENABLED; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.create_cq.opcode = SSSRAID_ADMIN_CREATE_CQ; + admin_cmd.create_cq.prp1 = cpu_to_le64(cqinfo->cq_dma_addr); + admin_cmd.create_cq.cqid = cpu_to_le16(qidx); + admin_cmd.create_cq.qsize = cpu_to_le16(sqinfo->q_depth - 1); + admin_cmd.create_cq.cq_flags = cpu_to_le16(flags); + admin_cmd.create_cq.irq_vector = cpu_to_le16(qidx); + + return sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); +} + +static int sssraid_create_io_cq(struct sssraid_dev *hdev, u16 qidx) +{ + int retval; + struct sssraid_cqueue *cqinfo = hdev->cqinfo + qidx; + u16 midx = qidx; + + retval = sssraid_create_cq(hdev, qidx); + if (retval) + return retval; + + /* + * cqinfo initialization at sssraid_init_queue + */ + hdev->intr_info[midx].cqinfo = cqinfo; + + return retval; +} + +static int sssraid_create_sq(struct sssraid_dev *hdev, u16 qidx) +{ + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + struct sssraid_admin_command admin_cmd; + int flags = SSSRAID_QUEUE_PHYS_CONTIG; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.create_sq.opcode = SSSRAID_ADMIN_CREATE_SQ; + admin_cmd.create_sq.prp1 = cpu_to_le64(sqinfo->sq_dma_addr); + admin_cmd.create_sq.sqid = cpu_to_le16(qidx); + admin_cmd.create_sq.qsize = cpu_to_le16(sqinfo->q_depth - 1); + admin_cmd.create_sq.sq_flags = cpu_to_le16(flags); + admin_cmd.create_sq.cqid = cpu_to_le16(qidx); + + return sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); +} + +static int sssraid_create_io_sq(struct sssraid_dev *hdev, u16 qidx) +{ + return sssraid_create_sq(hdev, qidx); +} + +int sssraid_get_dev_list(struct sssraid_dev *hdev, struct sssraid_dev_info *devices) +{ + u32 nd = le32_to_cpu(hdev->ctrl_info->nd); + struct sssraid_admin_command admin_cmd; + struct sssraid_dev_list *list_buf; + dma_addr_t data_dma = 0; + u32 i, idx, hdid, ndev; + int ret = 0; + + list_buf = dma_alloc_coherent(&hdev->pdev->dev, PAGE_SIZE, &data_dma, GFP_KERNEL); + if (!list_buf) + return -ENOMEM; + + for (idx = 0; idx < nd;) { + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.get_info.opcode = SSSRAID_ADMIN_GET_INFO; + admin_cmd.get_info.type = SSSRAID_GET_INFO_DEV_LIST; + admin_cmd.get_info.cdw11 = cpu_to_le32(idx); + admin_cmd.common.dptr.prp1 = cpu_to_le64(data_dma); + + ret = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); + + if (ret) { + ioc_err(hdev, "Get device list failed, nd: %u, idx: %u, ret: %d\n", + nd, idx, ret); + goto out; + } + ndev = le32_to_cpu(list_buf->dev_num); + + ioc_info(hdev, "ndev numbers: %u\n", ndev); + + for (i = 0; i < ndev; i++) { + hdid = le32_to_cpu(list_buf->devices[i].hdid); + ioc_info(hdev, "list_buf->devices[%d], hdid: %u target: %d, channel: %d, lun: %d, attr[0x%x]\n", + i, hdid, le16_to_cpu(list_buf->devices[i].target), + list_buf->devices[i].channel, + list_buf->devices[i].lun, + list_buf->devices[i].attr); + if (hdid > nd || hdid == 0) { + ioc_err(hdev, "err, hdid[%d] invalid\n", hdid); + continue; + } + memcpy(&devices[hdid - 1], &list_buf->devices[i], + sizeof(struct sssraid_dev_info)); + } + idx += ndev; + + if (ndev < MAX_DEV_ENTRY_PER_PAGE_4K) + break; + } + +out: + dma_free_coherent(&hdev->pdev->dev, PAGE_SIZE, list_buf, data_dma); + return ret; +} + +/* send abort command by admin queue temporary */ +int sssraid_send_abort_cmd(struct sssraid_dev *hdev, u32 hdid, u16 qidx, u16 cid) +{ + struct sssraid_admin_command admin_cmd; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.abort.opcode = SSSRAID_ADMIN_ABORT_CMD; + admin_cmd.abort.hdid = cpu_to_le32(hdid); + admin_cmd.abort.sqid = cpu_to_le16(qidx); + admin_cmd.abort.cid = cpu_to_le16(cid); + + return sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); +} + +/* send reset command by admin quueue temporary */ +int sssraid_send_reset_cmd(struct sssraid_dev *hdev, u8 type, u32 hdid) +{ + struct sssraid_admin_command admin_cmd; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.reset.opcode = SSSRAID_ADMIN_RESET; + admin_cmd.reset.hdid = cpu_to_le32(hdid); + admin_cmd.reset.type = type; + + return sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); +} + +static int sssraid_delete_queue(struct sssraid_dev *hdev, u8 op, u16 qidx) +{ + struct sssraid_admin_command admin_cmd; + int retval; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.delete_queue.opcode = op; + admin_cmd.delete_queue.qid = cpu_to_le16(qidx); + + retval = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); + + if (retval) + ioc_err(hdev, "Delete %s:[%d] failed\n", + (op == SSSRAID_ADMIN_DELETE_CQ) ? "cq" : "sq", qidx); + + return retval; +} + +static int sssraid_delete_cq(struct sssraid_dev *hdev, u16 qidx) +{ + return sssraid_delete_queue(hdev, SSSRAID_ADMIN_DELETE_CQ, qidx); +} + +void sssraid_adm_timeout(struct sssraid_dev *hdev, struct sssraid_cmd *cmd) +{ + /* command may be returned because controller reset */ + if (READ_ONCE(cmd->state) == SSSRAID_CMD_COMPLETE) + return; + + if (!sssraid_change_host_state(hdev, SSSRAID_RESETTING)) { + ioc_info(hdev, "can't change to reset state\n"); + return; + } + sssraid_soft_reset_handler(hdev); +} + +static int sssraid_create_io_qpair(struct sssraid_dev *hdev, u16 qidx) +{ + int retval; + + retval = sssraid_create_io_cq(hdev, qidx); + if (retval) + return retval; + + retval = sssraid_create_io_sq(hdev, qidx); + if (retval) + goto delete_cq; + + /* intr_info.msix_index substitute cq_vector */ + + /* io interrupt registry: + * not here, put above + */ + + sssraid_init_queue(hdev, qidx); + + return 0; + +delete_cq: + sssraid_delete_cq(hdev, qidx); + + return retval; +} + +static int sssraid_setup_io_qpair(struct sssraid_dev *hdev) +{ + u32 i, num_queues; + int retval = 0; + + num_queues = min(hdev->intr_info_count, hdev->init_done_queue_cnt - 1); + for (i = 1; i <= num_queues; i++) { + retval = sssraid_create_io_qpair(hdev, i); + if (retval) { + ioc_err(hdev, "Create queue[%d] failed\n", i); + break; + } + } + + ioc_info(hdev, "init_done_queue_cnt[%d], intr_info_count[%d] num_queues[%d]", + hdev->init_done_queue_cnt, /*hdev->online_queues,*/ + hdev->intr_info_count, num_queues); + + return retval >= 0 ? 0 : retval; +} + +static int sssraid_alloc_ioq_ptcmds(struct sssraid_dev *hdev) +{ + int i; + int ptnum = SSSRAID_NR_IOQ_PTCMDS; + + INIT_LIST_HEAD(&hdev->ioq_pt_list); + spin_lock_init(&hdev->ioq_pt_lock); + + hdev->ioq_ptcmds = kcalloc_node(ptnum, sizeof(struct sssraid_cmd), + GFP_KERNEL, hdev->numa_node); + + if (!hdev->ioq_ptcmds) { + ioc_err(hdev, "Alloc ioq_ptcmds failed\n"); + return -ENOMEM; + } + + for (i = 0; i < ptnum; i++) { + hdev->ioq_ptcmds[i].qid = i / SSSRAID_PTCMDS_PERQ + 1; + hdev->ioq_ptcmds[i].cid = i % SSSRAID_PTCMDS_PERQ + SSSRAID_IO_BLK_MQ_DEPTH; + list_add_tail(&(hdev->ioq_ptcmds[i].list), &hdev->ioq_pt_list); + } + + ioc_info(hdev, "Alloc ioq_ptcmds success, ptnum[%d]\n", ptnum); + + return 0; +} + +int sssraid_send_event_ack(struct sssraid_dev *hdev, u8 event, + u32 event_ctx, u16 cid) +{ + /* event,event_ctx no use at this time */ + struct sssraid_squeue *sqinfo = &hdev->sqinfo[0]; + struct sssraid_admin_command admin_cmd; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.common.opcode = SSSRAID_ADMIN_ASYNC_EVENT; + admin_cmd.common.command_id = cpu_to_le16(cid); + + sssraid_submit_cmd(sqinfo, &admin_cmd); + ioc_info(hdev, "send aen, cid[%d]\n", cid); + + return 0; +} + +static void sssraid_handle_aen_notice(struct sssraid_dev *hdev, u32 result) +{ + switch ((result & 0xff00) >> 8) { + case SSSRAID_AEN_DEV_CHANGED: + sssraid_scan_disk(hdev); + break; + case SSSRAID_AEN_FW_ACT_START: + ioc_info(hdev, "fw activation starting\n"); + break; + case SSSRAID_AEN_HOST_PROBING: + break; + default: + ioc_warn(hdev, "async event result %08x\n", result); + } +} + +static void sssraid_handle_aen_vs(struct sssraid_dev *hdev, u32 result, u32 result1) +{ + switch ((result & 0xff00) >> 8) { + case SSSRAID_AEN_TIMESYN: + sssraid_configure_timestamp(hdev); + break; + case SSSRAID_AEN_FW_ACT_FINISH: + ioc_info(hdev, "fw activation finish\n"); + if (sssraid_init_ctrl_info(hdev)) + ioc_err(hdev, "get ctrl info failed after fw act\n"); + break; + case SSSRAID_AEN_EVENT_MIN ... SSSRAID_AEN_EVENT_MAX: + ioc_info(hdev, "rcv card event[%d], param1[0x%x] param2[0x%x]\n", + (result & 0xff00) >> 8, result, result1); + break; + default: + ioc_warn(hdev, "async event result: 0x%x\n", result); + } +} + +static inline void sssraid_send_all_aen(struct sssraid_dev *hdev) +{ + u16 i; + + for (i = 0; i < hdev->ctrl_info->aerl; i++) + sssraid_send_event_ack(hdev, 0, 0, i + SSSRAID_AQ_BLK_MQ_DEPTH); +} + +static int sssraid_dev_list_init(struct sssraid_dev *hdev) +{ + u32 nd = le32_to_cpu(hdev->ctrl_info->nd); + + hdev->devices = kzalloc_node(nd * sizeof(struct sssraid_dev_info), + GFP_KERNEL, hdev->numa_node); + if (!hdev->devices) + return -ENOMEM; + + return 0; +} + +int sssraid_configure_timestamp(struct sssraid_dev *hdev) +{ + __le64 ts; + int retval; + + ts = cpu_to_le64(ktime_to_ms(ktime_get_real())); + retval = sssraid_set_features(hdev, SSSRAID_FEAT_TIMESTAMP, 0, &ts, sizeof(ts), NULL); + + if (retval) + ioc_err(hdev, "set timestamp failed: %d\n", retval); + return retval; +} + +int sssraid_init_ioc(struct sssraid_dev *hdev, u8 re_init) +{ + int retval = 0; + int i; + u32 nr_ioqs, bar_size; + + if (!re_init) { + hdev->cpu_count = num_online_cpus(); + + retval = sssraid_alloc_resources(hdev); + if (retval) { + ioc_err(hdev, "Failed to alloc resources:error %d\n", + retval); + goto out_nocleanup; + } + } + + /* reset need re-setup */ + retval = sssraid_setup_resources(hdev); + if (retval) { + ioc_err(hdev, "Failed to setup resources:error %d\n", + retval); + goto out_failed; + } + + if (!re_init) { + retval = sssraid_alloc_admin_cmds(hdev); + if (retval) { + ioc_err(hdev, "Failed to alloc admin cmds:error %d\n", + retval); + goto out_failed; + } + /* put here: + * alloc admin queue + */ + retval = sssraid_alloc_qpair(hdev, 0, SSSRAID_AQ_DEPTH); + if (retval) { + ioc_err(hdev, "Failed to alloc admin queue:error %d\n", + retval); + goto out_failed; + } + } + + retval = sssraid_setup_admin_qpair(hdev); + if (retval) { + ioc_err(hdev, "Failed to setup admin queue:error %d\n", + retval); + goto out_failed; + } + + /* 1. unregister all interrupt + * 2. admin interrupt registry + */ + retval = sssraid_setup_isr(hdev, 1); + if (retval) { + ioc_err(hdev, "Failed to setup ISR error %d\n", + retval); + goto out_failed; + } + + retval = sssraid_init_ctrl_info(hdev); + if (retval) { + ioc_err(hdev, "Failed to get ctrl info error %d\n", + retval); + goto out_failed; + } + + nr_ioqs = hdev->cpu_count; + retval = sssraid_set_queue_cnt(hdev, &nr_ioqs); + if (retval) { + ioc_err(hdev, "Failed to set queue cnt error %d\n", + retval); + goto out_failed; + } + + hdev->before_affinity_msix_cnt = nr_ioqs + 1; + + /* 1. unregister all interrupt + * 2. admin interrupt re-registry + * 3. io interrupt registry + */ + retval = sssraid_setup_isr(hdev, 0); + if (retval) { + ioc_err(hdev, "Failed to re-setup ISR, error %d\n", + retval); + goto out_failed; + } + + /* remap */ + bar_size = SSSRAID_REG_DBS + ((nr_ioqs + 1) * 8 * hdev->db_stride); + retval = sssraid_remap_bar(hdev, bar_size); + if (retval) { + ioc_err(hdev, "Failed to re-map bar, error %d\n", + retval); + goto out_failed; + } + hdev->sqinfo[0].q_db = hdev->dbs; + + /* num_vecs no sense, abandon */ + + if (!re_init) { + for (i = hdev->init_done_queue_cnt; i <= hdev->intr_info_count; i++) { + retval = sssraid_alloc_qpair(hdev, i, hdev->ioq_depth); + if (retval) { + ioc_err(hdev, "Failed to alloc io queue:error %d\n", + retval); + goto out_failed; + } + } + ioc_info(hdev, "intr_info_count: %d, init_done_queue_cnt: %d, ioq_depth: %d\n", + hdev->intr_info_count, hdev->init_done_queue_cnt, hdev->ioq_depth); + } + + retval = sssraid_setup_io_qpair(hdev); + if (retval) { + ioc_err(hdev, "Failed to setup io qpair, error %d\n", + retval); + goto out_failed; + } + + if (!re_init) { + retval = sssraid_alloc_ioq_ptcmds(hdev); + if (retval) { + ioc_err(hdev, "Failed to alloc ioq ptcmds, error %d\n", + retval); + goto out_failed; + } + } + + sssraid_send_all_aen(hdev); + + if (!re_init) { + retval = sssraid_dev_list_init(hdev); + if (retval) { + ioc_err(hdev, "Failed to init device list, error %d\n", + retval); + goto out_failed; + } + + retval = sssraid_configure_timestamp(hdev); + if (retval) { + ioc_err(hdev, "Failed to configure timestamp, error %d\n", + retval); + goto out_failed; + } + } + + return retval; + +out_failed: + sssraid_cleanup_ioc(hdev, re_init); +out_nocleanup: + return retval; +} + +void sssraid_cleanup_resources(struct sssraid_dev *hdev) +{ + struct pci_dev *pdev = hdev->pdev; + + sssraid_cleanup_isr(hdev); + + if (hdev->bar) { + iounmap(hdev->bar); + hdev->bar = NULL; + } + + if (pci_is_enabled(pdev)) { + pci_disable_pcie_error_reporting(pdev); + pci_release_mem_regions(pdev); + pci_disable_device(pdev); + } +} + +static void sssraid_free_ioq_ptcmds(struct sssraid_dev *hdev) +{ + kfree(hdev->ioq_ptcmds); + hdev->ioq_ptcmds = NULL; + + INIT_LIST_HEAD(&hdev->ioq_pt_list); +} + +static void sssraid_delete_io_queues(struct sssraid_dev *hdev) +{ + u16 queues = hdev->init_done_queue_cnt - 1; + u8 opcode = SSSRAID_ADMIN_DELETE_SQ; + u16 i, pass; + + if (!pci_device_is_present(hdev->pdev)) { + ioc_err(hdev, "pci_device is not present, skip disable io queues\n"); + return; + } + + if (hdev->init_done_queue_cnt < 2) { + ioc_err(hdev, "err, io queue has been delete\n"); + return; + } + + for (pass = 0; pass < 2; pass++) { + for (i = queues; i > 0; i--) + if (sssraid_delete_queue(hdev, opcode, i)) + break; + + opcode = SSSRAID_ADMIN_DELETE_CQ; + } +} + +void sssraid_complete_aen(struct sssraid_dev *hdev, struct sssraid_completion *cqe) +{ + u32 result = le32_to_cpu(cqe->result); + + ioc_info(hdev, "rcv aen, cid[%d], status[0x%x], result[0x%x]\n", + le16_to_cpu(cqe->cmd_id), le16_to_cpu(cqe->status) >> 1, result); + + /* + * The response to event moved from this func. + * sssraid_send_aen changed to name sssraid_send_event_ack + */ + + if ((le16_to_cpu(cqe->status) >> 1) != SSSRAID_SC_SUCCESS) + return; + switch (result & 0x7) { + case SSSRAID_AEN_NOTICE: + sssraid_handle_aen_notice(hdev, result); + break; + case SSSRAID_AEN_VS: + sssraid_handle_aen_vs(hdev, result, le32_to_cpu(cqe->result1)); + break; + default: + ioc_warn(hdev, "Unsupported async event type: %u\n", + result & 0x7); + break; + } +} + +void sssraid_free_iod_res(struct sssraid_dev *hdev, struct sssraid_iod *iod) +{ + const int last_prp = hdev->page_size / sizeof(__le64) - 1; + dma_addr_t dma_addr, next_dma_addr; + struct sssraid_sgl_desc *sg_list; + __le64 *prp_list; + void *addr; + int i; + + dma_addr = iod->first_dma; + if (iod->npages == 0) + dma_pool_free(iod->sqinfo->prp_small_pool, sssraid_iod_list(iod)[0], dma_addr); + + for (i = 0; i < iod->npages; i++) { + addr = sssraid_iod_list(iod)[i]; + + if (iod->use_sgl) { + sg_list = addr; + next_dma_addr = + le64_to_cpu((sg_list[SGES_PER_PAGE - 1]).addr); + } else { + prp_list = addr; + next_dma_addr = le64_to_cpu(prp_list[last_prp]); + } + + dma_pool_free(hdev->prp_page_pool, addr, dma_addr); + dma_addr = next_dma_addr; + } + + iod->sense = NULL; + iod->npages = -1; +} + +static void sssraid_complete_ioq_sync_cmnd(struct sssraid_dev *hdev, u16 qidx, + struct sssraid_completion *cqe) +{ + struct sssraid_cmd *ptcmd; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + + ptcmd = hdev->ioq_ptcmds + (sqinfo->qidx - 1) * SSSRAID_PTCMDS_PERQ + + le16_to_cpu(cqe->cmd_id) - SSSRAID_IO_BLK_MQ_DEPTH; + + ptcmd->status = le16_to_cpu(cqe->status) >> 1; + ptcmd->result0 = le32_to_cpu(cqe->result); + ptcmd->result1 = le32_to_cpu(cqe->result1); + + complete(&ptcmd->cmd_done); +} + +static void sssraid_complete_ioq_cmnd(struct sssraid_dev *hdev, u16 qidx, + struct sssraid_completion *cqe) +{ + struct sssraid_squeue *sqinfo = &hdev->sqinfo[qidx]; + + struct blk_mq_tags *tags; + struct scsi_cmnd *scmd; + struct sssraid_iod *iod; + struct request *req; + unsigned long elapsed; + + tags = hdev->shost->tag_set.tags[sqinfo->qidx - 1]; + + req = blk_mq_tag_to_rq(tags, le16_to_cpu(cqe->cmd_id)); + if (unlikely(!req || !blk_mq_request_started(req))) { + ioc_warn(hdev, "Invalid id %d completed on queue %d\n", + le16_to_cpu(cqe->cmd_id), sqinfo->qidx); + return; + } + + scmd = blk_mq_rq_to_pdu(req); + iod = scsi_cmd_priv(scmd); + + elapsed = jiffies - scmd->jiffies_at_alloc; + dbgprint(hdev, "cid[%d] qidx[%d] finish IO cost %3ld.%3ld seconds\n", + le16_to_cpu(cqe->cmd_id), sqinfo->qidx, elapsed / HZ, elapsed % HZ); + + if (cmpxchg(&iod->state, SSSRAID_CMD_IN_FLIGHT, SSSRAID_CMD_COMPLETE) != + SSSRAID_CMD_IN_FLIGHT) { + ioc_warn(hdev, "cid[%d] qidx[%d] enters abnormal handler, cost %3ld.%3ld seconds\n", + le16_to_cpu(cqe->cmd_id), sqinfo->qidx, elapsed / HZ, elapsed % HZ); + WRITE_ONCE(iod->state, SSSRAID_CMD_TMO_COMPLETE); + + if (iod->nsge) { + iod->nsge = 0; + scsi_dma_unmap(scmd); + } + sssraid_free_iod_res(hdev, iod); + + return; + } + + sssraid_map_status(iod, scmd, cqe); + if (iod->nsge) { + iod->nsge = 0; + scsi_dma_unmap(scmd); + } + sssraid_free_iod_res(hdev, iod); + scmd->scsi_done(scmd); +} + +static void sssraid_process_admin_cq(struct sssraid_dev *hdev, + struct sssraid_squeue *sqinfo, + struct sssraid_completion *cqe) +{ + struct sssraid_fwevt *fwevt = NULL; + u16 cid = le16_to_cpu(cqe->cmd_id), sz; + + if (likely(cid < SSSRAID_AQ_BLK_MQ_DEPTH)) + sssraid_complete_adminq_cmnd(hdev, sqinfo->qidx, cqe); + else { + sz = sizeof(*cqe); + fwevt = sssraid_alloc_fwevt(sz); + if (!fwevt) { + ioc_info(hdev, "%s :failure at %s:%d/%s()!\n", + __func__, __FILE__, __LINE__, __func__); + return; + } + + memcpy(fwevt->event_data, cqe, sz); + fwevt->hdev = hdev; + fwevt->event_id = 0; /* evt_type:0 */ + fwevt->send_ack = 1; /* ack_req:1 */ + fwevt->process_evt = 1; /* process_evt_bh:1 */ + fwevt->evt_ctx = 0; /* 0 */ + sssraid_fwevt_add_to_list(hdev, fwevt); + } +} + +static void sssraid_process_io_cq(struct sssraid_dev *hdev, + struct sssraid_squeue *sqinfo, + struct sssraid_completion *cqe) +{ + u16 cid = le16_to_cpu(cqe->cmd_id); + + if (likely(cid < SSSRAID_IO_BLK_MQ_DEPTH)) + sssraid_complete_ioq_cmnd(hdev, sqinfo->qidx, cqe); + else /* io sync handle */ + sssraid_complete_ioq_sync_cmnd(hdev, sqinfo->qidx, cqe); +} + +static inline void sssraid_handle_cqe(struct sssraid_dev *hdev, u16 mdix, u16 didx) +{ + struct sssraid_cqueue *cqinfo = &hdev->cqinfo[mdix]; + struct sssraid_squeue *sqinfo = &hdev->sqinfo[mdix]; + + struct sssraid_completion *cqe = &cqinfo->cqes[didx]; + u16 cid = le16_to_cpu(cqe->cmd_id); + + if (unlikely(cid >= sqinfo->q_depth)) { + ioc_err(hdev, "Invalid command id[%d] completed on queue %d\n", + cid, cqe->sq_id); + return; + } + + dbgprint(hdev, "cid[%d] mdix[%d], result[0x%x], sq_id[%d], status[0x%x]\n", + cid, sqinfo->qidx, le32_to_cpu(cqe->result), + le16_to_cpu(cqe->sq_id), le16_to_cpu(cqe->status)); + + if (!mdix) /* admin */ + sssraid_process_admin_cq(hdev, sqinfo, cqe); + else /* io */ + sssraid_process_io_cq(hdev, sqinfo, cqe); +} + +void sssraid_complete_cqes(struct sssraid_dev *hdev, u16 midx, u16 start, u16 end) +{ + struct sssraid_squeue *sqinfo = &hdev->sqinfo[midx]; + + while (start != end) { + sssraid_handle_cqe(hdev, midx, start); + if (++start == sqinfo->q_depth) + start = 0; + } +} + +static void sssraid_disable_admin_queue(struct sssraid_dev *hdev, bool shutdown) +{ + struct sssraid_cqueue *adm_cqinfo = &hdev->cqinfo[0]; + u16 start, end; + + if (pci_device_is_present(hdev->pdev)) { + if (shutdown) + sssraid_shutdown_ctrl(hdev); + else + sssraid_disable_ctrl(hdev); + } + + if (hdev->init_done_queue_cnt == 0) { + ioc_err(hdev, "err, admin queue has been delete\n"); + return; + } + + spin_lock_irq(&adm_cqinfo->cq_lock); + sssraid_process_cq(hdev, 0, &start, &end, -1); + spin_unlock_irq(&adm_cqinfo->cq_lock); + sssraid_complete_cqes(hdev, 0, start, end); +} + +static void sssraid_free_all_queues(struct sssraid_dev *hdev) +{ + int i; + struct sssraid_cqueue *cqinfo; + struct sssraid_squeue *sqinfo; + + for (i = 0; i < hdev->init_done_queue_cnt; i++) { + cqinfo = &hdev->cqinfo[i]; + sqinfo = &hdev->sqinfo[i]; + dma_free_coherent(&hdev->pdev->dev, CQ_SIZE(sqinfo->q_depth), + (void *)cqinfo->cqes, cqinfo->cq_dma_addr); + dma_free_coherent(&hdev->pdev->dev, SQ_SIZE(sqinfo->qidx, sqinfo->q_depth), + sqinfo->sq_cmds, sqinfo->sq_dma_addr); + dma_free_coherent(&hdev->pdev->dev, SENSE_SIZE(sqinfo->q_depth), + sqinfo->sense, sqinfo->sense_dma_addr); + } + + hdev->init_done_queue_cnt = 0; +} + +static void sssraid_free_admin_cmds(struct sssraid_dev *hdev) +{ + kfree(hdev->adm_cmds); + hdev->adm_cmds = NULL; + INIT_LIST_HEAD(&hdev->adm_cmd_list); +} + +static void sssraid_free_resources(struct sssraid_dev *hdev) +{ + sssraid_free_admin_cmds(hdev); + kfree(hdev->sqinfo); + kfree(hdev->cqinfo); + sssraid_destroy_dma_pools(hdev); + kfree(hdev->ctrl_info); +} + +void sssraid_cleanup_ioc(struct sssraid_dev *hdev, u8 re_init) +{ + if (!re_init) + sssraid_free_ioq_ptcmds(hdev); + + sssraid_delete_io_queues(hdev); + sssraid_disable_admin_queue(hdev, !re_init); + + if (!re_init) + sssraid_free_all_queues(hdev); + + sssraid_ioc_disable_intr(hdev); + sssraid_cleanup_resources(hdev); + + if (!re_init) + sssraid_free_resources(hdev); + +} + +int sssraid_soft_reset_handler(struct sssraid_dev *hdev) +{ + int retval = 0; + + if (hdev->state != SSSRAID_RESETTING) { + ioc_err(hdev, "err, host is not reset state\n"); + return retval; + } + + ioc_info(hdev, "enter host reset\n"); + + sssraid_ioc_disable_intr(hdev); + sssraid_cleanup_fwevt_list(hdev); + + /* realize above here: + * sssraid_dev_disable -> sssraid_back_all_io + */ + sssraid_back_all_io(hdev); + /* + * realize sssraid_dev_disable, + * i.e. sssraid_cleanup_ioc(1) + */ + if (hdev->ctrl_config & SSSRAID_CC_ENABLE) { + ioc_info(hdev, "start dev_disable\n"); + sssraid_cleanup_ioc(hdev, 1); + } + + retval = sssraid_init_ioc(hdev, 1); + if (retval) { + ioc_err(hdev, "init ioc fail.\n"); + return retval; + } + + sssraid_change_host_state(hdev, SSSRAID_LIVE); + + return retval; +} diff --git a/drivers/scsi/sssraid/sssraid_os.c b/drivers/scsi/sssraid/sssraid_os.c new file mode 100644 index 000000000000..16d2b94d78f3 --- /dev/null +++ b/drivers/scsi/sssraid/sssraid_os.c @@ -0,0 +1,2337 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright(c) 2022 3SNIC Information Technology, Ltd */ + +/* 3SNIC RAID SSSXXX Series Linux Driver */ + +#include <linux/version.h> +#include <linux/pci.h> +#include <linux/aer.h> +#include <linux/module.h> +#include <linux/ioport.h> +#include <linux/device.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/cdev.h> +#include <linux/sysfs.h> +#include <linux/gfp.h> +#include <linux/types.h> +#include <linux/ratelimit.h> +#include <linux/debugfs.h> +#include <linux/blkdev.h> +#include <linux/bsg-lib.h> +#include <linux/sort.h> + +#include <scsi/scsi.h> +#include <scsi/scsi_cmnd.h> +#include <scsi/scsi_device.h> +#include <scsi/scsi_host.h> +#include <scsi/scsi_transport.h> +#include <scsi/scsi_dbg.h> + +#include <linux/unaligned/be_byteshift.h> +#include <linux/unaligned/le_byteshift.h> +#include <linux/once.h> +#include <linux/sched/signal.h> +#include <linux/io-64-nonatomic-lo-hi.h> + +#include "sssraid.h" +#include "sssraid_debug.h" + +u32 admin_tmout = 60; +module_param(admin_tmout, uint, 0644); +MODULE_PARM_DESC(admin_tmout, "admin commands timeout (seconds)"); + +static u32 scmd_tmout_rawdisk = 180; +module_param(scmd_tmout_rawdisk, uint, 0644); +MODULE_PARM_DESC(scmd_tmout_rawdisk, "scsi commands timeout for rawdisk(seconds)"); + +static u32 scmd_tmout_vd = 180; +module_param(scmd_tmout_vd, uint, 0644); +MODULE_PARM_DESC(scmd_tmout_vd, "scsi commands timeout for vd(seconds)"); + +static int ioq_depth_set(const char *val, const struct kernel_param *kp); +static const struct kernel_param_ops ioq_depth_ops = { + .set = ioq_depth_set, + .get = param_get_uint, +}; + +u32 io_queue_depth = 1024; +module_param_cb(io_queue_depth, &ioq_depth_ops, &io_queue_depth, 0644); +MODULE_PARM_DESC(io_queue_depth, "set io queue depth, should >= 2"); + +static int logging_level_set(const char *val, const struct kernel_param *kp) +{ + u8 n = 0; + int ret; + + ret = kstrtou8(val, 10, &n); + if (ret != 0) + return -EINVAL; + + return param_set_byte(val, kp); +} + +static const struct kernel_param_ops logging_level_ops = { + .set = logging_level_set, + .get = param_get_byte, +}; + +static unsigned char logging_level; +module_param_cb(logging_level, &logging_level_ops, &logging_level, 0644); +MODULE_PARM_DESC(logging_level, "set log level, default zero for switch off"); + +static int small_pool_num_set(const char *val, const struct kernel_param *kp) +{ + u8 n = 0; + int ret; + + ret = kstrtou8(val, 10, &n); + if (ret != 0) + return -EINVAL; + if (n > MAX_SMALL_POOL_NUM) + n = MAX_SMALL_POOL_NUM; + if (n < 1) + n = 1; + *((u8 *)kp->arg) = n; + + return 0; +} + +static const struct kernel_param_ops small_pool_num_ops = { + .set = small_pool_num_set, + .get = param_get_byte, +}; + +/* Small pools are used to save PRP for small IOs.It was + * found that the spinlock of a single pool conflicts a + * lot with multiple CPUs.So multiple pools are introduced + * to reduce the conflictions. + */ +unsigned char small_pool_num = 4; +module_param_cb(small_pool_num, &small_pool_num_ops, &small_pool_num, 0644); +MODULE_PARM_DESC(small_pool_num, "set prp small pool num, default 4, MAX 16"); + +//static struct class *sssraid_class; + +enum FW_STAT_CODE { + FW_STAT_OK = 0, + FW_STAT_NEED_CHECK, + FW_STAT_ERROR, + FW_STAT_EP_PCIE_ERROR, + FW_STAT_NAC_DMA_ERROR, + FW_STAT_ABORTED, + FW_STAT_NEED_RETRY +}; + +static const char * const raid_levels[] = {"0", "1", "5", "6", "10", "50", "60", "NA"}; + +static const char * const raid_states[] = { + "NA", "NORMAL", "FAULT", "DEGRADE", "NOT_FORMATTED", "FORMATTING", "SANITIZING", + "INITIALIZING", "INITIALIZE_FAIL", "DELETING", "DELETE_FAIL", "WRITE_PROTECT" +}; + +static int ioq_depth_set(const char *val, const struct kernel_param *kp) +{ + int n = 0; + int ret; + + ret = kstrtoint(val, 10, &n); + if (ret != 0 || n < 2) + return -EINVAL; + + return param_set_int(val, kp); +} + +/* + * common + */ +static struct class *sssraid_class; + +struct sssraid_fwevt *sssraid_alloc_fwevt(int len) +{ + struct sssraid_fwevt *fwevt; + + fwevt = kzalloc(sizeof(*fwevt) + len, GFP_ATOMIC); + if (!fwevt) + return NULL; + + kref_init(&fwevt->ref_count); + return fwevt; + +} + +static void sssraid_fwevt_free(struct kref *r) +{ + kfree(container_of(r, struct sssraid_fwevt, ref_count)); +} + +static void sssraid_fwevt_get(struct sssraid_fwevt *fwevt) +{ + kref_get(&fwevt->ref_count); +} + +static void sssraid_fwevt_put(struct sssraid_fwevt *fwevt) +{ + kref_put(&fwevt->ref_count, sssraid_fwevt_free); +} + +static void sssraid_fwevt_del_from_list(struct sssraid_dev *hdev, + struct sssraid_fwevt *fwevt) +{ + unsigned long flags; + + spin_lock_irqsave(&hdev->fwevt_lock, flags); + if (!list_empty(&fwevt->list)) { + list_del_init(&fwevt->list); + /* + * Put fwevt reference count after + * removing it from fwevt_list + */ + sssraid_fwevt_put(fwevt); + } + spin_unlock_irqrestore(&hdev->fwevt_lock, flags); +} + +static void sssraid_fwevt_bh(struct sssraid_dev *hdev, + struct sssraid_fwevt *fwevt) +{ + struct sssraid_completion *cqe; + + hdev->current_event = fwevt; + sssraid_fwevt_del_from_list(hdev, fwevt); + + cqe = (struct sssraid_completion *)fwevt->event_data; + + if (!fwevt->process_evt) + goto evt_ack; + + sssraid_complete_aen(hdev, cqe); + +evt_ack: + /* event response put here: event has been handled. */ + sssraid_send_event_ack(hdev, fwevt->event_id, + fwevt->evt_ctx, le16_to_cpu(cqe->cmd_id)); + sssraid_fwevt_put(fwevt); + hdev->current_event = NULL; +} + +static void sssraid_fwevt_worker(struct work_struct *work) +{ + struct sssraid_fwevt *fwevt = container_of(work, struct sssraid_fwevt, + work); + sssraid_fwevt_bh(fwevt->hdev, fwevt); + /* + * Put fwevt reference count after + * dequeuing it from worker queue + */ + sssraid_fwevt_put(fwevt); +} + +void sssraid_fwevt_add_to_list(struct sssraid_dev *hdev, + struct sssraid_fwevt *fwevt) +{ + unsigned long flags; + + if (!hdev->fwevt_worker_thread) + return; + + spin_lock_irqsave(&hdev->fwevt_lock, flags); + /* get fwevt reference count while adding it to fwevt_list */ + sssraid_fwevt_get(fwevt); + INIT_LIST_HEAD(&fwevt->list); + list_add_tail(&fwevt->list, &hdev->fwevt_list); + INIT_WORK(&fwevt->work, sssraid_fwevt_worker); + /* get fwevt reference count while enqueueing it to worker queue */ + sssraid_fwevt_get(fwevt); + queue_work(hdev->fwevt_worker_thread, &fwevt->work); + spin_unlock_irqrestore(&hdev->fwevt_lock, flags); +} + +static struct sssraid_fwevt *sssraid_dequeue_fwevt( + struct sssraid_dev *hdev) +{ + unsigned long flags; + struct sssraid_fwevt *fwevt = NULL; + + spin_lock_irqsave(&hdev->fwevt_lock, flags); + if (!list_empty(&hdev->fwevt_list)) { + fwevt = list_first_entry(&hdev->fwevt_list, + struct sssraid_fwevt, list); + list_del_init(&fwevt->list); + /* + * Put fwevt reference count after + * removing it from fwevt_list + */ + sssraid_fwevt_put(fwevt); + } + spin_unlock_irqrestore(&hdev->fwevt_lock, flags); + + return fwevt; +} + +void sssraid_cleanup_fwevt_list(struct sssraid_dev *hdev) +{ + struct sssraid_fwevt *fwevt = NULL; + + if ((list_empty(&hdev->fwevt_list) && !hdev->current_event) || + !hdev->fwevt_worker_thread) + return; + + while ((fwevt = sssraid_dequeue_fwevt(hdev)) || + (fwevt = hdev->current_event)) { + /* + * Wait on the fwevt to complete. If this returns 1, then + * the event was never executed, and we need a put for the + * reference the work had on the fwevt. + * + * If it did execute, we wait for it to finish, and the put will + * happen from sssraid_process_fwevt() + */ + if (cancel_work_sync(&fwevt->work)) { + /* + * Put fwevt reference count after + * dequeuing it from worker queue + */ + sssraid_fwevt_put(fwevt); + /* + * Put fwevt reference count to neutralize + * kref_init increment + */ + sssraid_fwevt_put(fwevt); + } + } +} + +/* + * common 1 + */ +static int sssraid_npages_prp(struct sssraid_dev *hdev) +{ + u32 size = 1U << ((hdev->ctrl_info->mdts) * 1U) << 12; + u32 nprps = DIV_ROUND_UP(size + hdev->page_size, hdev->page_size); + + return DIV_ROUND_UP(PRP_ENTRY_SIZE * nprps, hdev->page_size - PRP_ENTRY_SIZE); +} + +static int sssraid_npages_sgl(struct sssraid_dev *hdev) +{ + u32 nsge = le16_to_cpu(hdev->ctrl_info->max_num_sge); + + return DIV_ROUND_UP(nsge * sizeof(struct sssraid_sgl_desc), hdev->page_size); +} + +static u32 sssraid_cmd_size(struct sssraid_dev *hdev) +{ + u32 alloc_size = sizeof(__le64 *) * max(sssraid_npages_prp(hdev), sssraid_npages_sgl(hdev)); + + ioc_info(hdev, "iod size: %lu, alloc_size: %u\n", + sizeof(struct sssraid_iod), alloc_size); + + return sizeof(struct sssraid_iod) + alloc_size; +} + +static int sssraid_setup_prps(struct sssraid_dev *hdev, struct sssraid_iod *iod) +{ + struct scatterlist *sg = iod->sg; + u64 dma_addr = sg_dma_address(sg); + int dma_len = sg_dma_len(sg); + __le64 *prp_list, *old_prp_list; + int page_size = hdev->page_size; + int offset = dma_addr & (page_size - 1); + void **list = sssraid_iod_list(iod); + int length = iod->length; + struct dma_pool *pool; + dma_addr_t prp_dma; + int nprps, i; + + length -= (page_size - offset); + if (length <= 0) { + iod->first_dma = 0; + return 0; + } + + dma_len -= (page_size - offset); + if (dma_len) { + dma_addr += (page_size - offset); + } else { + sg = sg_next(sg); + dma_addr = sg_dma_address(sg); + dma_len = sg_dma_len(sg); + } + + if (length <= page_size) { + iod->first_dma = dma_addr; + return 0; + } + + nprps = DIV_ROUND_UP(length, page_size); + if (nprps <= (SMALL_POOL_SIZE / PRP_ENTRY_SIZE)) { + pool = iod->sqinfo->prp_small_pool; + iod->npages = 0; + } else { + pool = hdev->prp_page_pool; + iod->npages = 1; + } + + prp_list = dma_pool_alloc(pool, GFP_ATOMIC, &prp_dma); + if (!prp_list) { + dev_err_ratelimited(&hdev->pdev->dev, "Allocate first prp_list memory failed\n"); + iod->first_dma = dma_addr; + iod->npages = -1; + return -ENOMEM; + } + list[0] = prp_list; + iod->first_dma = prp_dma; + i = 0; + for (;;) { + if (i == page_size / PRP_ENTRY_SIZE) { + old_prp_list = prp_list; + + prp_list = dma_pool_alloc(pool, GFP_ATOMIC, &prp_dma); + if (!prp_list) { + dev_err_ratelimited(&hdev->pdev->dev, "Allocate %dth prp_list memory failed\n", + iod->npages + 1); + return -ENOMEM; + } + list[iod->npages++] = prp_list; + prp_list[0] = old_prp_list[i - 1]; + old_prp_list[i - 1] = cpu_to_le64(prp_dma); + i = 1; + } + prp_list[i++] = cpu_to_le64(dma_addr); + dma_len -= page_size; + dma_addr += page_size; + length -= page_size; + if (length <= 0) + break; + if (dma_len > 0) + continue; + if (unlikely(dma_len < 0)) + goto bad_sgl; + sg = sg_next(sg); + dma_addr = sg_dma_address(sg); + dma_len = sg_dma_len(sg); + } + + return 0; + +bad_sgl: + ioc_err(hdev, "Setup prps, invalid SGL for payload: %d nents: %d\n", + iod->length, iod->nsge); + return -EIO; +} + +static inline bool sssraid_is_rw_scmd(struct scsi_cmnd *scmd) +{ + switch (scmd->cmnd[0]) { + case READ_6: + case READ_10: + case READ_12: + case READ_16: + case WRITE_6: + case WRITE_10: + case WRITE_12: + case WRITE_16: + return true; + default: + return false; + } +} + +static bool sssraid_is_prp(struct sssraid_dev *hdev, struct scsi_cmnd *scmd, u32 nsge) +{ + struct scatterlist *sg = scsi_sglist(scmd); + u32 page_mask = hdev->page_size - 1; + bool is_prp = true; + int i = 0; + + scsi_for_each_sg(scmd, sg, nsge, i) { + if (i != 0 && i != nsge - 1) { + if ((sg_dma_len(sg) & page_mask) || + (sg_dma_address(sg) & page_mask)) { + is_prp = false; + break; + } + } + + if (nsge > 1 && i == 0) { + if ((sg_dma_address(sg) + sg_dma_len(sg)) & page_mask) { + is_prp = false; + break; + } + } + + if (nsge > 1 && i == (nsge - 1)) { + if (sg_dma_address(sg) & page_mask) { + is_prp = false; + break; + } + } + } + + return is_prp; +} + +static void sssraid_sgl_set_data(struct sssraid_sgl_desc *sge, struct scatterlist *sg) +{ + sge->addr = cpu_to_le64(sg_dma_address(sg)); + sge->length = cpu_to_le32(sg_dma_len(sg)); + sge->type = SSSRAID_SGL_FMT_DATA_DESC << 4; +} + +static void sssraid_sgl_set_seg(struct sssraid_sgl_desc *sge, dma_addr_t dma_addr, int entries) +{ + sge->addr = cpu_to_le64(dma_addr); + if (entries <= SGES_PER_PAGE) { + sge->length = cpu_to_le32(entries * sizeof(*sge)); + sge->type = SSSRAID_SGL_FMT_LAST_SEG_DESC << 4; + } else { + sge->length = cpu_to_le32(PAGE_SIZE); + sge->type = SSSRAID_SGL_FMT_SEG_DESC << 4; + } +} + + +static int sssraid_setup_ioq_cmd_sgl(struct sssraid_dev *hdev, + struct scsi_cmnd *scmd, struct sssraid_ioq_command *ioq_cmd, + struct sssraid_iod *iod) +{ + struct sssraid_sgl_desc *sg_list, *link, *old_sg_list; + struct scatterlist *sg = scsi_sglist(scmd); + void **list = sssraid_iod_list(iod); + struct dma_pool *pool; + int nsge = iod->nsge; + dma_addr_t sgl_dma; + int i = 0; + + ioq_cmd->common.flags |= SSSRAID_CMD_FLAG_SGL_METABUF; + + if (nsge == 1) { + sssraid_sgl_set_data(&ioq_cmd->common.dptr.sgl, sg); + return 0; + } + + if (nsge <= (SMALL_POOL_SIZE / sizeof(struct sssraid_sgl_desc))) { + pool = iod->sqinfo->prp_small_pool; + iod->npages = 0; + } else { + pool = hdev->prp_page_pool; + iod->npages = 1; + } + + sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &sgl_dma); + if (!sg_list) { + dev_err_ratelimited(&hdev->pdev->dev, "Allocate first sgl_list failed\n"); + iod->npages = -1; + return -ENOMEM; + } + + list[0] = sg_list; + iod->first_dma = sgl_dma; + sssraid_sgl_set_seg(&ioq_cmd->common.dptr.sgl, sgl_dma, nsge); + do { + if (i == SGES_PER_PAGE) { + old_sg_list = sg_list; + link = &old_sg_list[SGES_PER_PAGE - 1]; + + sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &sgl_dma); + if (!sg_list) { + dev_err_ratelimited(&hdev->pdev->dev, "Allocate %dth sgl_list failed\n", + iod->npages + 1); + return -ENOMEM; + } + list[iod->npages++] = sg_list; + + i = 0; + memcpy(&sg_list[i++], link, sizeof(*link)); + sssraid_sgl_set_seg(link, sgl_dma, nsge); + } + + sssraid_sgl_set_data(&sg_list[i++], sg); + sg = sg_next(sg); + } while (--nsge > 0); + + return 0; +} + +static void sssraid_shost_init(struct sssraid_dev *hdev) +{ + struct pci_dev *pdev = hdev->pdev; + u8 domain, bus; + u32 dev_func; + + domain = pci_domain_nr(pdev->bus); + bus = pdev->bus->number; + dev_func = pdev->devfn; + + hdev->shost->nr_hw_queues = hdev->init_done_queue_cnt - 1; + hdev->shost->can_queue = (hdev->ioq_depth - SSSRAID_PTCMDS_PERQ); + + hdev->shost->sg_tablesize = le16_to_cpu(hdev->ctrl_info->max_num_sge); + /* 512B per sector */ + hdev->shost->max_sectors = (1U << ((hdev->ctrl_info->mdts) * 1U) << 12) / 512; + hdev->shost->cmd_per_lun = MAX_CMD_PER_DEV; + hdev->shost->max_channel = le16_to_cpu(hdev->ctrl_info->max_channel) - 1; + hdev->shost->max_id = le32_to_cpu(hdev->ctrl_info->max_tgt_id); + hdev->shost->max_lun = le16_to_cpu(hdev->ctrl_info->max_lun); + + hdev->shost->this_id = -1; + hdev->shost->unique_id = (domain << 16) | (bus << 8) | dev_func; + hdev->shost->max_cmd_len = MAX_CDB_LEN; + hdev->shost->hostt->cmd_size = sssraid_cmd_size(hdev); +} + +static inline void sssraid_get_tag_from_scmd(struct scsi_cmnd *scmd, u16 *qidx, u16 *cid) +{ + u32 tag = blk_mq_unique_tag(scmd->request); + + *qidx = blk_mq_unique_tag_to_hwq(tag) + 1; + *cid = blk_mq_unique_tag_to_tag(tag); +} + +static inline uint32_t get_unaligned_be24(const uint8_t *const p) +{ + return get_unaligned_be32(p - 1) & 0xffffffU; +} + +static int sssraid_setup_rw_cmd(struct sssraid_dev *hdev, + struct sssraid_rw_command *rw, + struct scsi_cmnd *scmd) +{ + u32 start_lba_lo, start_lba_hi; + u32 datalength = 0; + u16 control = 0; + + start_lba_lo = 0; + start_lba_hi = 0; + + if (scmd->sc_data_direction == DMA_TO_DEVICE) { + rw->opcode = SSSRAID_CMD_WRITE; + } else if (scmd->sc_data_direction == DMA_FROM_DEVICE) { + rw->opcode = SSSRAID_CMD_READ; + } else { + ioc_err(hdev, "Invalid RW_IO for unsupported data direction: %d\n", + scmd->sc_data_direction); + WARN_ON(1); + return -EINVAL; + } + + /* 6-byte READ(0x08) or WRITE(0x0A) cdb */ + if (scmd->cmd_len == 6) { + datalength = (u32)(scmd->cmnd[4] == 0 ? + IO_6_DEFAULT_TX_LEN : scmd->cmnd[4]); + start_lba_lo = (u32)get_unaligned_be24(&scmd->cmnd[1]); + + start_lba_lo &= 0x1FFFFF; + } + + /* 10-byte READ(0x28) or WRITE(0x2A) cdb */ + else if (scmd->cmd_len == 10) { + datalength = (u32)get_unaligned_be16(&scmd->cmnd[7]); + start_lba_lo = get_unaligned_be32(&scmd->cmnd[2]); + + if (scmd->cmnd[1] & FUA_MASK) + control |= SSSRAID_RW_FUA; + } + + /* 12-byte READ(0xA8) or WRITE(0xAA) cdb */ + else if (scmd->cmd_len == 12) { + datalength = get_unaligned_be32(&scmd->cmnd[6]); + start_lba_lo = get_unaligned_be32(&scmd->cmnd[2]); + + if (scmd->cmnd[1] & FUA_MASK) + control |= SSSRAID_RW_FUA; + } + /* 16-byte READ(0x88) or WRITE(0x8A) cdb */ + else if (scmd->cmd_len == 16) { + datalength = get_unaligned_be32(&scmd->cmnd[10]); + start_lba_lo = get_unaligned_be32(&scmd->cmnd[6]); + start_lba_hi = get_unaligned_be32(&scmd->cmnd[2]); + + if (scmd->cmnd[1] & FUA_MASK) + control |= SSSRAID_RW_FUA; + } + + if (unlikely(datalength > U16_MAX || datalength == 0)) { + ioc_err(hdev, "Invalid IO for illegal transfer data length: %u\n", datalength); + WARN_ON(1); + return -EINVAL; + } + + rw->slba = cpu_to_le64(((u64)start_lba_hi << 32) | start_lba_lo); + /* 0base for nlb */ + rw->nlb = cpu_to_le16((u16)(datalength - 1)); + rw->control = cpu_to_le16(control); + + return 0; +} + +static int sssraid_setup_nonio_cmd(struct sssraid_dev *hdev, + struct sssraid_scsi_nonio *scsi_nonio, struct scsi_cmnd *scmd) +{ + scsi_nonio->buffer_len = cpu_to_le32(scsi_bufflen(scmd)); + + switch (scmd->sc_data_direction) { + case DMA_NONE: + scsi_nonio->opcode = SSSRAID_CMD_NONIO_NONE; + break; + case DMA_TO_DEVICE: + scsi_nonio->opcode = SSSRAID_CMD_NONIO_TODEV; + break; + case DMA_FROM_DEVICE: + scsi_nonio->opcode = SSSRAID_CMD_NONIO_FROMDEV; + break; + default: + ioc_err(hdev, "Invalid NON_IO for unsupported data direction: %d\n", + scmd->sc_data_direction); + WARN_ON(1); + return -EINVAL; + } + + return 0; +} + +static int sssraid_setup_ioq_cmd(struct sssraid_dev *hdev, + struct sssraid_ioq_command *ioq_cmd, struct scsi_cmnd *scmd) +{ + memcpy(ioq_cmd->common.cdb, scmd->cmnd, scmd->cmd_len); + ioq_cmd->common.cdb_len = scmd->cmd_len; + + if (sssraid_is_rw_scmd(scmd)) + return sssraid_setup_rw_cmd(hdev, &ioq_cmd->rw, scmd); + else + return sssraid_setup_nonio_cmd(hdev, &ioq_cmd->scsi_nonio, scmd); +} + +static inline void sssraid_init_iod(struct sssraid_iod *iod) +{ + iod->nsge = 0; + iod->npages = -1; + iod->use_sgl = false; + WRITE_ONCE(iod->state, SSSRAID_CMD_IDLE); +} + +int sssraid_io_map_data(struct sssraid_dev *hdev, struct sssraid_iod *iod, + struct scsi_cmnd *scmd, struct sssraid_ioq_command *ioq_cmd) +{ + int retval; + + retval = scsi_dma_map(scmd); + if (unlikely(retval < 0)) + return retval; + iod->nsge = retval; + /* No data to DMA, it may be scsi no-rw command */ + if (unlikely(iod->nsge == 0)) + return 0; + + iod->length = scsi_bufflen(scmd); + iod->sg = scsi_sglist(scmd); + iod->use_sgl = !sssraid_is_prp(hdev, scmd, iod->nsge); + + if (iod->use_sgl) { + retval = sssraid_setup_ioq_cmd_sgl(hdev, scmd, ioq_cmd, iod); + } else { + retval = sssraid_setup_prps(hdev, iod); + ioq_cmd->common.dptr.prp1 = + cpu_to_le64(sg_dma_address(iod->sg)); + ioq_cmd->common.dptr.prp2 = cpu_to_le64(iod->first_dma); + } + + if (retval) + scsi_dma_unmap(scmd); + + return retval; +} + +void sssraid_map_status(struct sssraid_iod *iod, struct scsi_cmnd *scmd, + struct sssraid_completion *cqe) +{ + struct sssraid_dev *hdev = iod->sqinfo->hdev; + + scsi_set_resid(scmd, 0); + + switch ((le16_to_cpu(cqe->status) >> 1) & 0x7f) { + case FW_STAT_OK: + set_host_byte(scmd, DID_OK); + break; + case FW_STAT_NEED_CHECK: + set_host_byte(scmd, DID_OK); + scmd->result |= le16_to_cpu(cqe->status) >> 8; + if (scmd->result & SAM_STAT_CHECK_CONDITION) { + memset(scmd->sense_buffer, 0, SCSI_SENSE_BUFFERSIZE); + memcpy(scmd->sense_buffer, iod->sense, SCSI_SENSE_BUFFERSIZE); + scmd->result = (scmd->result & 0x00ffffff) | (DRIVER_SENSE << 24); + } + break; + case FW_STAT_ABORTED: + set_host_byte(scmd, DID_ABORT); + break; + case FW_STAT_NEED_RETRY: + set_host_byte(scmd, DID_REQUEUE); + break; + default: + set_host_byte(scmd, DID_BAD_TARGET); + ioc_warn(hdev, "cid[%d] qid[%d] bad status[0x%x]\n", + le16_to_cpu(cqe->cmd_id), le16_to_cpu(cqe->sq_id), + le16_to_cpu(cqe->status)); + break; + } +} + + +struct sssraid_cmd *sssraid_get_cmd(struct sssraid_dev *hdev, enum sssraid_cmd_type type) +{ + struct sssraid_cmd *cmd = NULL; + unsigned long flags; + struct list_head *head = &hdev->adm_cmd_list; + spinlock_t *slock = &hdev->adm_cmd_lock; + + if (type == SSSRAID_CMD_IOPT) { + head = &hdev->ioq_pt_list; + slock = &hdev->ioq_pt_lock; + } + + spin_lock_irqsave(slock, flags); + if (list_empty(head)) { + spin_unlock_irqrestore(slock, flags); + ioc_err(hdev, "err, cmd[%d] list empty\n", type); + return NULL; + } + cmd = list_entry(head->next, struct sssraid_cmd, list); + list_del_init(&cmd->list); + spin_unlock_irqrestore(slock, flags); + + WRITE_ONCE(cmd->state, SSSRAID_CMD_IN_FLIGHT); + + return cmd; +} + +static int sssraid_add_device(struct sssraid_dev *hdev, struct sssraid_dev_info *device) +{ + struct Scsi_Host *shost = hdev->shost; + struct scsi_device *sdev; + + ioc_info(hdev, "add device, hdid: %u target: %d, channel: %d, lun: %d, attr[0x%x]\n", + le32_to_cpu(device->hdid), le16_to_cpu(device->target), + device->channel, device->lun, device->attr); + + sdev = scsi_device_lookup(shost, device->channel, le16_to_cpu(device->target), 0); + if (sdev) { + ioc_warn(hdev, "Device is already exist, channel: %d, target_id: %d, lun: %d\n", + device->channel, le16_to_cpu(device->target), 0); + scsi_device_put(sdev); + return -EEXIST; + } + scsi_add_device(shost, device->channel, le16_to_cpu(device->target), 0); + return 0; +} + +static int sssraid_rescan_device(struct sssraid_dev *hdev, struct sssraid_dev_info *device) +{ + struct Scsi_Host *shost = hdev->shost; + struct scsi_device *sdev; + + ioc_info(hdev, "rescan device, hdid: %u target: %d, channel: %d, lun: %d, attr[0x%x]\n", + le32_to_cpu(device->hdid), le16_to_cpu(device->target), + device->channel, device->lun, device->attr); + + sdev = scsi_device_lookup(shost, device->channel, le16_to_cpu(device->target), 0); + if (!sdev) { + ioc_warn(hdev, "device is not exit rescan it, channel: %d, target_id: %d, lun: %d\n", + device->channel, le16_to_cpu(device->target), 0); + return -ENODEV; + } + + scsi_rescan_device(&sdev->sdev_gendev); + scsi_device_put(sdev); + return 0; +} + +static int sssraid_remove_device(struct sssraid_dev *hdev, struct sssraid_dev_info *org_device) +{ + struct Scsi_Host *shost = hdev->shost; + struct scsi_device *sdev; + + ioc_info(hdev, "remove device, hdid: %u target: %d, channel: %d, lun: %d, attr[0x%x]\n", + le32_to_cpu(org_device->hdid), le16_to_cpu(org_device->target), + org_device->channel, org_device->lun, org_device->attr); + + sdev = scsi_device_lookup(shost, org_device->channel, le16_to_cpu(org_device->target), 0); + if (!sdev) { + ioc_warn(hdev, "device is not exit remove it, channel: %d, target_id: %d, lun: %d\n", + org_device->channel, le16_to_cpu(org_device->target), 0); + return -ENODEV; + } + + scsi_remove_device(sdev); + scsi_device_put(sdev); + return 0; +} + +static int luntarget_cmp_func(const void *l, const void *r) +{ + const struct sssraid_dev_info *ln = l; + const struct sssraid_dev_info *rn = r; + int l_attr = SSSRAID_DEV_INFO_ATTR_BOOT(ln->attr); + int r_attr = SSSRAID_DEV_INFO_ATTR_BOOT(rn->attr); + + /* boot first */ + if (l_attr != r_attr) + return (r_attr - l_attr); + + if (ln->channel == rn->channel) + return le16_to_cpu(ln->target) - le16_to_cpu(rn->target); + + return ln->channel - rn->channel; +} + +void sssraid_scan_disk(struct sssraid_dev *hdev) +{ + struct sssraid_dev_info *devices, *org_devices; + struct sssraid_dev_info *sortdevice; + u32 nd = le32_to_cpu(hdev->ctrl_info->nd); + u8 flag, org_flag; + int i, ret; + int count = 0; + + devices = kcalloc(nd, sizeof(struct sssraid_dev_info), GFP_KERNEL); + if (!devices) + return; + + sortdevice = kcalloc(nd, sizeof(struct sssraid_dev_info), GFP_KERNEL); + if (!sortdevice) + goto free_list; + + ret = sssraid_get_dev_list(hdev, devices); + if (ret) + goto free_all; + org_devices = hdev->devices; + for (i = 0; i < nd; i++) { + org_flag = org_devices[i].flag; + flag = devices[i].flag; + + dbgprint(hdev, "i: %d, org_flag: 0x%x, flag: 0x%x\n", i, org_flag, flag); + + if (SSSRAID_DEV_INFO_FLAG_VALID(flag)) { + if (!SSSRAID_DEV_INFO_FLAG_VALID(org_flag)) { + down_write(&hdev->devices_rwsem); + memcpy(&org_devices[i], &devices[i], + sizeof(struct sssraid_dev_info)); + memcpy(&sortdevice[count++], &devices[i], + sizeof(struct sssraid_dev_info)); + up_write(&hdev->devices_rwsem); + } else if (SSSRAID_DEV_INFO_FLAG_CHANGE(flag)) { + sssraid_rescan_device(hdev, &devices[i]); + } + } else { + if (SSSRAID_DEV_INFO_FLAG_VALID(org_flag)) { + down_write(&hdev->devices_rwsem); + org_devices[i].flag &= 0xfe; + up_write(&hdev->devices_rwsem); + sssraid_remove_device(hdev, &org_devices[i]); + } + } + } + + ioc_info(hdev, "scan work add device count = %d\n", count); + + sort(sortdevice, count, sizeof(sortdevice[0]), luntarget_cmp_func, NULL); + + for (i = 0; i < count; i++) + sssraid_add_device(hdev, &sortdevice[i]); + +free_all: + kfree(sortdevice); +free_list: + kfree(devices); +} + +static int sssraid_wait_abnl_cmd_done(struct sssraid_iod *iod) +{ + u16 times = 0; + + do { + if (READ_ONCE(iod->state) == SSSRAID_CMD_TMO_COMPLETE) + break; + msleep(500); + times++; + } while (times <= SSSRAID_WAIT_ABNL_CMD_TIMEOUT); + + /* wait command completion timeout after abort/reset success */ + if (times >= SSSRAID_WAIT_ABNL_CMD_TIMEOUT) + return -ETIMEDOUT; + + return 0; +} + +static bool sssraid_check_scmd_completed(struct scsi_cmnd *scmd) +{ + struct sssraid_dev *hdev = shost_priv(scmd->device->host); + struct sssraid_iod *iod = scsi_cmd_priv(scmd); + struct sssraid_squeue *sqinfo; + u16 hwq, cid; + + sssraid_get_tag_from_scmd(scmd, &hwq, &cid); + sqinfo = &hdev->sqinfo[hwq]; + if (READ_ONCE(iod->state) == SSSRAID_CMD_COMPLETE || sssraid_poll_cq(hdev, hwq, cid)) { + ioc_warn(hdev, "cid[%d] qidx[%d] has been completed\n", + cid, sqinfo->qidx); + return true; + } + return false; +} + +static int sssraid_scsi_reset(struct scsi_cmnd *scmd, enum sssraid_scsi_rst_type rst) +{ + struct sssraid_dev *hdev = shost_priv(scmd->device->host); + struct sssraid_iod *iod = scsi_cmd_priv(scmd); + struct sssraid_sdev_hostdata *hostdata; + u16 hwq, cid; + int ret; + + scsi_print_command(scmd); + + if (hdev->state != SSSRAID_LIVE || !sssraid_wait_abnl_cmd_done(iod) || + sssraid_check_scmd_completed(scmd)) + return SUCCESS; + + hostdata = scmd->device->hostdata; + sssraid_get_tag_from_scmd(scmd, &hwq, &cid); + + ioc_warn(hdev, "cid[%d] qidx[%d] timeout, %s reset\n", cid, hwq, + rst ? "bus" : "target"); + ret = sssraid_send_reset_cmd(hdev, rst, hostdata->hdid); + if (ret == 0) { + ret = sssraid_wait_abnl_cmd_done(iod); + if (ret) { + ioc_warn(hdev, "cid[%d] qidx[%d] %s reset failed, not found\n", + cid, hwq, rst ? "bus" : "target"); + return FAILED; + } + + ioc_warn(hdev, "cid[%d] qidx[%d] %s reset success\n", cid, hwq, + rst ? "bus" : "target"); + return SUCCESS; + } + + ioc_warn(hdev, "cid[%d] qidx[%d] ret[%d] %s reset failed\n", cid, hwq, ret, + rst ? "bus" : "target"); + return FAILED; +} + +bool sssraid_change_host_state(struct sssraid_dev *hdev, enum sssraid_state newstate) +{ + unsigned long flags; + enum sssraid_state oldstate; + bool change = false; + + spin_lock_irqsave(&hdev->state_lock, flags); + + oldstate = hdev->state; + switch (newstate) { + case SSSRAID_LIVE: + switch (oldstate) { + case SSSRAID_NEW: + case SSSRAID_RESETTING: + change = true; + break; + default: + break; + } + break; + case SSSRAID_RESETTING: + switch (oldstate) { + case SSSRAID_LIVE: + change = true; + break; + default: + break; + } + break; + case SSSRAID_DELETING: + if (oldstate != SSSRAID_DELETING) + change = true; + break; + case SSSRAID_DEAD: + switch (oldstate) { + case SSSRAID_NEW: + case SSSRAID_LIVE: + case SSSRAID_RESETTING: + change = true; + break; + default: + break; + } + break; + default: + break; + } + if (change) + hdev->state = newstate; + spin_unlock_irqrestore(&hdev->state_lock, flags); + + ioc_info(hdev, "[%d]->[%d], change[%d]\n", oldstate, newstate, change); + + return change; +} + +static int sssraid_get_qd_by_disk(u8 attr) +{ + switch (SSSRAID_DEV_DISK_TYPE(attr)) { + case SSSRAID_SAS_HDD_VD: + case SSSRAID_SATA_HDD_VD: + return SSSRAID_HDD_VD_QD; + case SSSRAID_SAS_SSD_VD: + case SSSRAID_SATA_SSD_VD: + case SSSRAID_NVME_SSD_VD: + return SSSRAID_SSD_VD_QD; + case SSSRAID_SAS_HDD_PD: + case SSSRAID_SATA_HDD_PD: + return SSSRAID_HDD_PD_QD; + case SSSRAID_SAS_SSD_PD: + case SSSRAID_SATA_SSD_PD: + case SSSRAID_NVME_SSD_PD: + return SSSRAID_SSD_PD_QD; + default: + return MAX_CMD_PER_DEV; + } +} + +static int sssraid_match_dev(struct sssraid_dev *hdev, u16 idx, struct scsi_device *sdev) +{ + if (SSSRAID_DEV_INFO_FLAG_VALID(hdev->devices[idx].flag)) { + if (sdev->channel == hdev->devices[idx].channel && + sdev->id == le16_to_cpu(hdev->devices[idx].target) && + sdev->lun < hdev->devices[idx].lun) { + ioc_info(hdev, "Match device success, channel:target:lun[%d:%d:%d]\n", + hdev->devices[idx].channel, + hdev->devices[idx].target, + hdev->devices[idx].lun); + return 1; + } + } + + return 0; +} + +static int sssraid_bsg_map_data(struct sssraid_dev *hdev, struct bsg_job *job, + struct sssraid_admin_command *cmd) +{ + struct request *rq = blk_mq_rq_from_pdu(job); + struct sssraid_iod *iod = job->dd_data; + enum dma_data_direction dma_dir = rq_data_dir(rq) ? DMA_TO_DEVICE : DMA_FROM_DEVICE; + int ret = 0; + + iod->sg = job->request_payload.sg_list; + iod->nsge = job->request_payload.sg_cnt; + iod->length = job->request_payload.payload_len; + iod->use_sgl = false; + iod->npages = -1; + + if (!iod->nsge) + goto out; + + ret = dma_map_sg_attrs(&hdev->pdev->dev, iod->sg, iod->nsge, dma_dir, DMA_ATTR_NO_WARN); + if (!ret) + goto out; + + ret = sssraid_setup_prps(hdev, iod); + if (ret) + goto unmap; + + cmd->common.dptr.prp1 = cpu_to_le64(sg_dma_address(iod->sg)); + cmd->common.dptr.prp2 = cpu_to_le64(iod->first_dma); + + return 0; + +unmap: + dma_unmap_sg(&hdev->pdev->dev, iod->sg, iod->nsge, dma_dir); +out: + return ret; +} + +static void sssraid_bsg_unmap_data(struct sssraid_dev *hdev, struct bsg_job *job) +{ + struct request *rq = blk_mq_rq_from_pdu(job); + struct sssraid_iod *iod = job->dd_data; + enum dma_data_direction dma_dir = rq_data_dir(rq) ? DMA_TO_DEVICE : DMA_FROM_DEVICE; + + if (iod->nsge) + dma_unmap_sg(&hdev->pdev->dev, iod->sg, iod->nsge, dma_dir); + + sssraid_free_iod_res(hdev, iod); +} + +void sssraid_put_cmd(struct sssraid_dev *hdev, struct sssraid_cmd *cmd, + enum sssraid_cmd_type type) +{ + unsigned long flags; + struct list_head *head = &hdev->adm_cmd_list; + spinlock_t *slock = &hdev->adm_cmd_lock; + + if (type == SSSRAID_CMD_IOPT) { + head = &hdev->ioq_pt_list; + slock = &hdev->ioq_pt_lock; + } + + spin_lock_irqsave(slock, flags); + WRITE_ONCE(cmd->state, SSSRAID_CMD_IDLE); + list_add_tail(&cmd->list, head); + spin_unlock_irqrestore(slock, flags); +} + +static int sssraid_user_admin_cmd(struct sssraid_dev *hdev, struct bsg_job *job) +{ + struct sssraid_bsg_request *bsg_req = job->request; + struct sssraid_passthru_common_cmd *cmd = &(bsg_req->admcmd); + struct sssraid_admin_command admin_cmd; + u32 timeout = msecs_to_jiffies(cmd->timeout_ms); + u32 result[2] = {0}; + int status; + + if (hdev->state >= SSSRAID_RESETTING) { + ioc_err(hdev, "err, host state:[%d] is not right\n", + hdev->state); + return -EBUSY; + } + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.common.opcode = cmd->opcode; + admin_cmd.common.flags = cmd->flags; + admin_cmd.common.hdid = cpu_to_le32(cmd->nsid); + admin_cmd.common.cdw2[0] = cpu_to_le32(cmd->cdw2); + admin_cmd.common.cdw2[1] = cpu_to_le32(cmd->cdw3); + admin_cmd.common.cdw10 = cpu_to_le32(cmd->cdw10); + admin_cmd.common.cdw11 = cpu_to_le32(cmd->cdw11); + admin_cmd.common.cdw12 = cpu_to_le32(cmd->cdw12); + admin_cmd.common.cdw13 = cpu_to_le32(cmd->cdw13); + admin_cmd.common.cdw14 = cpu_to_le32(cmd->cdw14); + admin_cmd.common.cdw15 = cpu_to_le32(cmd->cdw15); + + status = sssraid_bsg_map_data(hdev, job, &admin_cmd); + if (status) { + ioc_err(hdev, "err, map data failed\n"); + return status; + } + + status = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, &result[0], &result[1], timeout); + if (status >= 0) { + job->reply_len = sizeof(result); + memcpy(job->reply, result, sizeof(result)); + } + + if (status) + ioc_info(hdev, "opcode[0x%x] subopcode[0x%x], status[0x%x] result0[0x%x] result1[0x%x]\n", + cmd->opcode, cmd->info_0.subopcode, status, result[0], result[1]); + + sssraid_bsg_unmap_data(hdev, job); + + return status; +} + +static int sssraid_submit_ioq_sync_cmd(struct sssraid_dev *hdev, struct sssraid_ioq_command *cmd, + u32 *result, u32 *reslen, u32 timeout) +{ + int ret; + dma_addr_t sense_dma; + struct sssraid_squeue *sqinfo; + void *sense_addr = NULL; + struct sssraid_cmd *pt_cmd = sssraid_get_cmd(hdev, SSSRAID_CMD_IOPT); + + if (!pt_cmd) { + ioc_err(hdev, "err, get sqinfo cmd failed\n"); + return -EFAULT; + } + + timeout = timeout ? timeout : ADMIN_TIMEOUT; + + init_completion(&pt_cmd->cmd_done); + + sqinfo = &hdev->sqinfo[pt_cmd->qid]; + ret = pt_cmd->cid * SCSI_SENSE_BUFFERSIZE; + sense_addr = sqinfo->sense + ret; + sense_dma = sqinfo->sense_dma_addr + ret; + + cmd->common.sense_addr = cpu_to_le64(sense_dma); + cmd->common.sense_len = cpu_to_le16(SCSI_SENSE_BUFFERSIZE); + cmd->common.command_id = cpu_to_le16(pt_cmd->cid); + + sssraid_submit_cmd(sqinfo, cmd); + + if (!wait_for_completion_timeout(&pt_cmd->cmd_done, timeout)) { + ioc_err(hdev, "cid[%d] qid[%d] timeout, opcode[0x%x] subopcode[0x%x]\n", + pt_cmd->cid, pt_cmd->qid, cmd->common.opcode, + (le32_to_cpu(cmd->common.cdw3[0]) & 0xffff)); + + /* reset controller if admin timeout */ + sssraid_adm_timeout(hdev, pt_cmd); + + sssraid_put_cmd(hdev, pt_cmd, SSSRAID_CMD_IOPT); + return -ETIME; + } + + if (result && reslen) { + if ((pt_cmd->status & 0x17f) == 0x101) { + memcpy(result, sense_addr, SCSI_SENSE_BUFFERSIZE); + *reslen = SCSI_SENSE_BUFFERSIZE; + } + } + + sssraid_put_cmd(hdev, pt_cmd, SSSRAID_CMD_IOPT); + + return pt_cmd->status; +} + +static int sssraid_user_ioq_cmd(struct sssraid_dev *hdev, struct bsg_job *job) +{ + struct sssraid_bsg_request *bsg_req = (struct sssraid_bsg_request *)(job->request); + struct sssraid_ioq_passthru_cmd *cmd = &(bsg_req->ioqcmd); + struct sssraid_ioq_command ioq_cmd; + int status = 0; + u32 timeout = msecs_to_jiffies(cmd->timeout_ms); + + if (cmd->data_len > IOQ_PT_DATA_LEN) { + ioc_err(hdev, "data len bigger than 4k\n"); + return -EFAULT; + } + + if (hdev->state != SSSRAID_LIVE) { + ioc_err(hdev, "err, host state:[%d] is not live\n", + hdev->state); + return -EBUSY; + } + + ioc_info(hdev, "opcode[0x%x] subopcode[0x%x] init, datalen[%d]\n", + cmd->opcode, cmd->info_1.subopcode, cmd->data_len); + + memset(&ioq_cmd, 0, sizeof(ioq_cmd)); + ioq_cmd.common.opcode = cmd->opcode; + ioq_cmd.common.flags = cmd->flags; + ioq_cmd.common.hdid = cpu_to_le32(cmd->nsid); + ioq_cmd.common.sense_len = cpu_to_le16(cmd->info_0.res_sense_len); + ioq_cmd.common.cdb_len = cmd->info_0.cdb_len; + ioq_cmd.common.rsvd2 = cmd->info_0.rsvd0; + ioq_cmd.common.cdw3[0] = cpu_to_le32(cmd->cdw3); + ioq_cmd.common.cdw3[1] = cpu_to_le32(cmd->cdw4); + ioq_cmd.common.cdw3[2] = cpu_to_le32(cmd->cdw5); + + ioq_cmd.common.cdw10[0] = cpu_to_le32(cmd->cdw10); + ioq_cmd.common.cdw10[1] = cpu_to_le32(cmd->cdw11); + ioq_cmd.common.cdw10[2] = cpu_to_le32(cmd->cdw12); + ioq_cmd.common.cdw10[3] = cpu_to_le32(cmd->cdw13); + ioq_cmd.common.cdw10[4] = cpu_to_le32(cmd->cdw14); + ioq_cmd.common.cdw10[5] = cpu_to_le32(cmd->data_len); + + memcpy(ioq_cmd.common.cdb, &cmd->cdw16, cmd->info_0.cdb_len); + + ioq_cmd.common.cdw26[0] = cpu_to_le32(cmd->cdw26[0]); + ioq_cmd.common.cdw26[1] = cpu_to_le32(cmd->cdw26[1]); + ioq_cmd.common.cdw26[2] = cpu_to_le32(cmd->cdw26[2]); + ioq_cmd.common.cdw26[3] = cpu_to_le32(cmd->cdw26[3]); + + status = sssraid_bsg_map_data(hdev, job, (struct sssraid_admin_command *)&ioq_cmd); + if (status) { + ioc_err(hdev, "err, map data failed\n"); + return status; + } + + status = sssraid_submit_ioq_sync_cmd(hdev, &ioq_cmd, job->reply, &job->reply_len, timeout); + if (status) + ioc_info(hdev, "opcode[0x%x] subopcode[0x%x], status[0x%x], reply_len[%d]\n", + cmd->opcode, cmd->info_1.subopcode, status, job->reply_len); + + sssraid_bsg_unmap_data(hdev, job); + + return status; +} + + +/* bsg dispatch user command */ +static int sssraid_bsg_host_dispatch(struct bsg_job *job) +{ + struct Scsi_Host *shost = dev_to_shost(job->dev); + struct sssraid_dev *hdev = shost_priv(shost); + struct request *rq = blk_mq_rq_from_pdu(job); + struct sssraid_bsg_request *bsg_req = job->request; + int ret = -ENOMSG; + + job->reply_len = 0; + + if (bsg_req == NULL || job->request_len != sizeof(struct sssraid_bsg_request)) { + bsg_job_done(job, ret, 0); + return 0; + } + dbgprint(hdev, "bsg msgcode[%d] msglen[%d] timeout[%d];" + "reqnsge[%d], reqlen[%d]\n", + bsg_req->msgcode, job->request_len, rq->timeout, + job->request_payload.sg_cnt, job->request_payload.payload_len); + + switch (bsg_req->msgcode) { + case SSSRAID_BSG_ADM: + ret = sssraid_user_admin_cmd(hdev, job); + break; + case SSSRAID_BSG_IOQ: + ret = sssraid_user_ioq_cmd(hdev, job); + break; + default: + ioc_info(hdev, "unsupport msgcode[%d]\n", bsg_req->msgcode); + break; + } + + if (ret > 0) + ret = ret | (ret << 8); + + bsg_job_done(job, ret, 0); + return 0; +} + +static inline void sssraid_remove_bsg(struct sssraid_dev *hdev) +{ + if (hdev->bsg_queue) { + bsg_unregister_queue(hdev->bsg_queue); + blk_cleanup_queue(hdev->bsg_queue); + } +} + +static void sssraid_back_fault_cqe(struct sssraid_squeue *sqinfo, struct sssraid_completion *cqe) +{ + struct sssraid_dev *hdev = sqinfo->hdev; + struct blk_mq_tags *tags; + struct scsi_cmnd *scmd; + struct sssraid_iod *iod; + struct request *req; + + tags = hdev->shost->tag_set.tags[sqinfo->qidx - 1]; + req = blk_mq_tag_to_rq(tags, le16_to_cpu(cqe->cmd_id)); + if (unlikely(!req || !blk_mq_request_started(req))) + return; + + scmd = blk_mq_rq_to_pdu(req); + iod = scsi_cmd_priv(scmd); + + if (READ_ONCE(iod->state) != SSSRAID_CMD_IN_FLIGHT && + READ_ONCE(iod->state) != SSSRAID_CMD_TIMEOUT) + return; + + WRITE_ONCE(iod->state, SSSRAID_CMD_TMO_COMPLETE); + set_host_byte(scmd, DID_NO_CONNECT); + if (iod->nsge) + scsi_dma_unmap(scmd); + sssraid_free_iod_res(hdev, iod); + scmd->scsi_done(scmd); + ioc_warn(hdev, "Back fault CQE, cid[%d] qidx[%d]\n", + le16_to_cpu(cqe->cmd_id), sqinfo->qidx); +} + +void sssraid_back_all_io(struct sssraid_dev *hdev) +{ + int i, j; + struct sssraid_squeue *sqinfo; + struct sssraid_completion cqe = { 0 }; + + scsi_block_requests(hdev->shost); + + for (i = 1; i <= hdev->shost->nr_hw_queues; i++) { + sqinfo = &hdev->sqinfo[i]; + for (j = 0; j < hdev->scsi_qd; j++) { + cqe.cmd_id = cpu_to_le16(j); + sssraid_back_fault_cqe(sqinfo, &cqe); + } + } + + scsi_unblock_requests(hdev->shost); + + j = SSSRAID_AQ_BLK_MQ_DEPTH; + for (i = 0; i < j; i++) { + if (READ_ONCE(hdev->adm_cmds[i].state) == SSSRAID_CMD_IN_FLIGHT) { + ioc_info(hdev, "Back adm, cid[%d]\n", i); + hdev->adm_cmds[i].status = 0xFFFF; + WRITE_ONCE(hdev->adm_cmds[i].state, SSSRAID_CMD_COMPLETE); + complete(&(hdev->adm_cmds[i].cmd_done)); + } + } + + j = SSSRAID_NR_IOQ_PTCMDS; + for (i = 0; i < j; i++) { + if (READ_ONCE(hdev->ioq_ptcmds[i].state) == SSSRAID_CMD_IN_FLIGHT) { + hdev->ioq_ptcmds[i].status = 0xFFFF; + WRITE_ONCE(hdev->ioq_ptcmds[i].state, SSSRAID_CMD_COMPLETE); + complete(&(hdev->ioq_ptcmds[i].cmd_done)); + } + } +} + +/** + * static struct scsi_host_template sssraid_driver_template + */ +static int sssraid_scan_finished(struct Scsi_Host *shost, + unsigned long time) +{ + struct sssraid_dev *hdev = shost_priv(shost); + + sssraid_scan_disk(hdev); + + return 1; +} + +/* eh_target_reset_handler call back */ +static int sssraid_eh_target_reset(struct scsi_cmnd *scmd) +{ + return sssraid_scsi_reset(scmd, SSSRAID_RESET_TARGET); +} + +/* eh_bus_reset_handler call back */ +static int sssraid_bus_reset_handler(struct scsi_cmnd *scmd) +{ + return sssraid_scsi_reset(scmd, SSSRAID_RESET_BUS); +} + +/* eh_host_reset_handler call back */ +static int sssraid_eh_host_reset(struct scsi_cmnd *scmd) +{ + u16 hwq, cid; + struct sssraid_dev *hdev = shost_priv(scmd->device->host); + + scsi_print_command(scmd); + if (hdev->state != SSSRAID_LIVE || sssraid_check_scmd_completed(scmd)) + return SUCCESS; + + sssraid_get_tag_from_scmd(scmd, &hwq, &cid); + ioc_warn(hdev, "cid[%d] qidx[%d] host reset\n", cid, hwq); + + /* It's useless: + * old code sssraid_reset_work_sync + * queue_work(reset_work) at first, + * then flush_work to synchronize. + */ + if (!sssraid_change_host_state(hdev, SSSRAID_RESETTING)) { + ioc_info(hdev, "can't change to reset state\n"); + return FAILED; + } + if (sssraid_soft_reset_handler(hdev)) { + ioc_warn(hdev, "cid[%d] qidx[%d] host reset failed\n", cid, hwq); + return FAILED; + } + + ioc_warn(hdev, "cid[%d] qidx[%d] host reset success\n", cid, hwq); + + return SUCCESS; +} + +/* host_reset call back */ +static int sssraid_sysfs_host_reset(struct Scsi_Host *shost, int reset_type) +{ + int ret; + struct sssraid_dev *hdev = shost_priv(shost); + + ioc_info(hdev, "start sysfs host reset cmd\n"); + if (!sssraid_change_host_state(hdev, SSSRAID_RESETTING)) { + ioc_info(hdev, "can't change to reset state\n"); + return -EBUSY; + } + ret = sssraid_soft_reset_handler(hdev); + ioc_info(hdev, "stop sysfs host reset cmd[%d]\n", ret); + + return ret; +} + +/* queuecommand call back */ +static int sssraid_qcmd(struct Scsi_Host *shost, + struct scsi_cmnd *scmd) +{ + struct sssraid_iod *iod = scsi_cmd_priv(scmd); + struct sssraid_dev *hdev = shost_priv(shost); + struct scsi_device *sdev = scmd->device; + struct sssraid_sdev_hostdata *hostdata = sdev->hostdata; + u16 hwq, cid; + struct sssraid_squeue *sq; + struct sssraid_ioq_command ioq_cmd; + int retval; + + if (unlikely(hdev->state == SSSRAID_RESETTING)) + return SCSI_MLQUEUE_HOST_BUSY; + + if (unlikely(hdev->state != SSSRAID_LIVE)) { + set_host_byte(scmd, DID_NO_CONNECT); + scmd->scsi_done(scmd); + return 0; + } + + if (unlikely(hdev->logging_level & SSSRAID_DEBUG)) + scsi_print_command(scmd); + + sssraid_get_tag_from_scmd(scmd, &hwq, &cid); + hostdata = sdev->hostdata; + sq = &hdev->sqinfo[hwq]; + + memset(&ioq_cmd, 0, sizeof(ioq_cmd)); + ioq_cmd.rw.hdid = cpu_to_le32(hostdata->hdid); + ioq_cmd.rw.command_id = cpu_to_le16(cid); + + retval = sssraid_setup_ioq_cmd(hdev, &ioq_cmd, scmd); + if (unlikely(retval)) { + set_host_byte(scmd, DID_ERROR); + scmd->scsi_done(scmd); + return 0; + } + + iod->sense = sq->sense + retval; + iod->sense_dma = sq->sense_dma_addr + retval; + ioq_cmd.common.sense_addr = cpu_to_le64(iod->sense_dma); + ioq_cmd.common.sense_len = cpu_to_le16(SCSI_SENSE_BUFFERSIZE); + + sssraid_init_iod(iod); + + iod->sqinfo = sq; + sssraid_io_map_data(hdev, iod, scmd, &ioq_cmd); + if (unlikely(retval)) { + ioc_err(hdev, "sssraid_io_map_data Err.\n"); + set_host_byte(scmd, DID_ERROR); + scmd->scsi_done(scmd); + retval = 0; + goto deinit_iod; + } + + WRITE_ONCE(iod->state, SSSRAID_CMD_IN_FLIGHT); + sssraid_submit_cmd(sq, &ioq_cmd); + + return 0; + +deinit_iod: + sssraid_free_iod_res(hdev, iod); + return retval; +} + +/* change_queue_depth call back: + * keep as old + */ + +/* slave_configure call back */ +static int sssraid_slave_configure(struct scsi_device *sdev) +{ + int qd = MAX_CMD_PER_DEV; + unsigned int timeout = scmd_tmout_rawdisk * HZ; + struct sssraid_dev *hdev = shost_priv(sdev->host); + struct sssraid_sdev_hostdata *hostdata = sdev->hostdata; + u32 max_sec = sdev->host->max_sectors; + + if (hostdata) { + if (SSSRAID_DEV_INFO_ATTR_VD(hostdata->attr)) + timeout = scmd_tmout_vd * HZ; + else if (SSSRAID_DEV_INFO_ATTR_RAWDISK(hostdata->attr)) + timeout = scmd_tmout_rawdisk * HZ; + max_sec = hostdata->max_io_kb << 1; + qd = sssraid_get_qd_by_disk(hostdata->attr); + } else { + ioc_err(hdev, "err, sdev->hostdata is null\n"); + } + + blk_queue_rq_timeout(sdev->request_queue, timeout); + sdev->eh_timeout = timeout; + scsi_change_queue_depth(sdev, qd); + + if ((max_sec == 0) || (max_sec > sdev->host->max_sectors)) + max_sec = sdev->host->max_sectors; + + blk_queue_max_hw_sectors(sdev->request_queue, max_sec); + + ioc_info(hdev, "sdev->channel:id:lun[%d:%d:%lld], scmd_timeout[%d]s, maxsec[%d]\n", + sdev->channel, sdev->id, sdev->lun, timeout / HZ, max_sec); + + return 0; +} + +/* slave_alloc call back */ +static int sssraid_slave_alloc(struct scsi_device *sdev) +{ + struct sssraid_sdev_hostdata *hostdata; + struct sssraid_dev *hdev; + u16 idx; + + hdev = shost_priv(sdev->host); + hostdata = kzalloc(sizeof(*hostdata), GFP_KERNEL); + if (!hostdata) { + ioc_err(hdev, "Alloc scsi host data memory failed\n"); + return -ENOMEM; + } + + down_read(&hdev->devices_rwsem); + for (idx = 0; idx < le32_to_cpu(hdev->ctrl_info->nd); idx++) { + if (sssraid_match_dev(hdev, idx, sdev)) + goto scan_host; + } + up_read(&hdev->devices_rwsem); + + kfree(hostdata); + return -ENXIO; + +scan_host: + hostdata->hdid = le32_to_cpu(hdev->devices[idx].hdid); + hostdata->max_io_kb = le16_to_cpu(hdev->devices[idx].max_io_kb); + hostdata->attr = hdev->devices[idx].attr; + hostdata->flag = hdev->devices[idx].flag; + hostdata->rg_id = 0xff; + sdev->hostdata = hostdata; + up_read(&hdev->devices_rwsem); + return 0; +} + +/* slave_destroy call back */ +static void sssraid_slave_destroy(struct scsi_device *sdev) +{ + kfree(sdev->hostdata); + sdev->hostdata = NULL; +} + +/* eh_timed_out call back */ +static enum blk_eh_timer_return sssraid_scmd_timeout(struct scsi_cmnd *scmd) +{ + struct sssraid_iod *iod = scsi_cmd_priv(scmd); + unsigned int timeout = scmd->device->request_queue->rq_timeout; + + if (sssraid_check_scmd_completed(scmd)) + goto out; + + if (time_after(jiffies, scmd->jiffies_at_alloc + timeout)) { + if (cmpxchg(&iod->state, SSSRAID_CMD_IN_FLIGHT, SSSRAID_CMD_TIMEOUT) == + SSSRAID_CMD_IN_FLIGHT) { + return BLK_EH_DONE; + } + } +out: + return BLK_EH_RESET_TIMER; +} + +/* eh_abort_handler call back */ +static int sssraid_abort_handler(struct scsi_cmnd *scmd) +{ + struct sssraid_dev *hdev = shost_priv(scmd->device->host); + struct sssraid_iod *iod = scsi_cmd_priv(scmd); + struct sssraid_sdev_hostdata *hostdata; + u16 hwq, cid; + int ret; + + scsi_print_command(scmd); + + if (hdev->state != SSSRAID_LIVE || !sssraid_wait_abnl_cmd_done(iod) || + sssraid_check_scmd_completed(scmd)) + return SUCCESS; + + hostdata = scmd->device->hostdata; + sssraid_get_tag_from_scmd(scmd, &hwq, &cid); + + ioc_warn(hdev, "cid[%d] qidx[%d] timeout, aborting\n", cid, hwq); + ret = sssraid_send_abort_cmd(hdev, hostdata->hdid, hwq, cid); + if (ret != -ETIME) { + ret = sssraid_wait_abnl_cmd_done(iod); + if (ret) { + ioc_warn(hdev, "cid[%d] qidx[%d] abort failed, not found\n", cid, hwq); + return FAILED; + } + ioc_warn(hdev, "cid[%d] qidx[%d] abort succ\n", cid, hwq); + return SUCCESS; + } + ioc_warn(hdev, "cid[%d] qidx[%d] abort failed, timeout\n", cid, hwq); + return FAILED; +} + +static ssize_t csts_pp_show(struct device *cdev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct sssraid_dev *hdev = shost_priv(shost); + int ret = -1; + + if (pci_device_is_present(hdev->pdev)) { + ret = (readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_PP_MASK); + ret >>= SSSRAID_CSTS_PP_SHIFT; + } + + return snprintf(buf, PAGE_SIZE, "%d\n", ret); +} + +static ssize_t csts_shst_show(struct device *cdev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct sssraid_dev *hdev = shost_priv(shost); + int ret = -1; + + if (pci_device_is_present(hdev->pdev)) { + ret = (readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_SHST_MASK); + ret >>= SSSRAID_CSTS_SHST_SHIFT; + } + + return snprintf(buf, PAGE_SIZE, "%d\n", ret); +} + +static ssize_t csts_cfs_show(struct device *cdev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct sssraid_dev *hdev = shost_priv(shost); + int ret = -1; + + if (pci_device_is_present(hdev->pdev)) { + ret = (readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_CFS_MASK); + ret >>= SSSRAID_CSTS_CFS_SHIFT; + } + + return snprintf(buf, PAGE_SIZE, "%d\n", ret); +} + +static ssize_t csts_rdy_show(struct device *cdev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct sssraid_dev *hdev = shost_priv(shost); + int ret = -1; + + if (pci_device_is_present(hdev->pdev)) + ret = (readl(hdev->bar + SSSRAID_REG_CSTS) & SSSRAID_CSTS_RDY); + + return snprintf(buf, PAGE_SIZE, "%d\n", ret); +} + +static ssize_t fw_version_show(struct device *cdev, struct device_attribute *attr, char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct sssraid_dev *hdev = shost_priv(shost); + + return snprintf(buf, PAGE_SIZE, "%s\n", hdev->ctrl_info->fr); +} + +static DEVICE_ATTR_RO(csts_pp); +static DEVICE_ATTR_RO(csts_shst); +static DEVICE_ATTR_RO(csts_cfs); +static DEVICE_ATTR_RO(csts_rdy); +static DEVICE_ATTR_RO(fw_version); + +static struct device_attribute *sssraid_host_attrs[] = { + &dev_attr_csts_pp, + &dev_attr_csts_shst, + &dev_attr_csts_cfs, + &dev_attr_csts_rdy, + &dev_attr_fw_version, + NULL, +}; + +static int sssraid_get_vd_info(struct sssraid_dev *hdev, struct sssraid_vd_info *vd_info, u16 vid) +{ + struct sssraid_admin_command admin_cmd; + u8 *data_ptr = NULL; + dma_addr_t data_dma = 0; + int ret; + + if (hdev->state >= SSSRAID_RESETTING) { + ioc_err(hdev, "err, host state[%d] is not right\n", hdev->state); + return -EBUSY; + } + + data_ptr = dma_alloc_coherent(&hdev->pdev->dev, PAGE_SIZE, &data_dma, GFP_KERNEL); + if (!data_ptr) + return -ENOMEM; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.usr_cmd.opcode = USR_CMD_READ; + admin_cmd.usr_cmd.info_0.subopcode = cpu_to_le16(USR_CMD_VDINFO); + admin_cmd.usr_cmd.info_1.data_len = cpu_to_le16(USR_CMD_RDLEN); + admin_cmd.usr_cmd.info_1.param_len = cpu_to_le16(VDINFO_PARAM_LEN); + admin_cmd.usr_cmd.cdw10 = cpu_to_le32(vid); + admin_cmd.common.dptr.prp1 = cpu_to_le64(data_dma); + + ret = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); + if (!ret) + memcpy(vd_info, data_ptr, sizeof(struct sssraid_vd_info)); + + dma_free_coherent(&hdev->pdev->dev, PAGE_SIZE, data_ptr, data_dma); + + return ret; +} + +static int sssraid_get_bgtask(struct sssraid_dev *hdev, struct sssraid_bgtask *bgtask) +{ + struct sssraid_admin_command admin_cmd; + u8 *data_ptr = NULL; + dma_addr_t data_dma = 0; + int ret; + + if (hdev->state >= SSSRAID_RESETTING) { + ioc_err(hdev, "err, host state[%d] is not right\n", hdev->state); + return -EBUSY; + } + + data_ptr = dma_alloc_coherent(&hdev->pdev->dev, PAGE_SIZE, &data_dma, GFP_KERNEL); + if (!data_ptr) + return -ENOMEM; + + memset(&admin_cmd, 0, sizeof(admin_cmd)); + admin_cmd.usr_cmd.opcode = USR_CMD_READ; + admin_cmd.usr_cmd.info_0.subopcode = cpu_to_le16(USR_CMD_BGTASK); + admin_cmd.usr_cmd.info_1.data_len = cpu_to_le16(USR_CMD_RDLEN); + admin_cmd.common.dptr.prp1 = cpu_to_le64(data_dma); + + ret = sssraid_submit_admin_sync_cmd(hdev, &admin_cmd, NULL, NULL, 0); + if (!ret) + memcpy(bgtask, data_ptr, sizeof(struct sssraid_bgtask)); + + dma_free_coherent(&hdev->pdev->dev, PAGE_SIZE, data_ptr, data_dma); + + return ret; +} + +static ssize_t raid_level_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct scsi_device *sdev; + struct sssraid_dev *hdev; + struct sssraid_vd_info *vd_info; + struct sssraid_sdev_hostdata *hostdata; + int ret; + + sdev = to_scsi_device(dev); + hdev = shost_priv(sdev->host); + hostdata = sdev->hostdata; + + vd_info = kmalloc(sizeof(*vd_info), GFP_KERNEL); + if (!vd_info || !SSSRAID_DEV_INFO_ATTR_VD(hostdata->attr)) + return snprintf(buf, PAGE_SIZE, "NA\n"); + + ret = sssraid_get_vd_info(hdev, vd_info, sdev->id); + if (ret) + vd_info->rg_level = ARRAY_SIZE(raid_levels) - 1; + + ret = (vd_info->rg_level < ARRAY_SIZE(raid_levels)) ? + vd_info->rg_level : (ARRAY_SIZE(raid_levels) - 1); + + kfree(vd_info); + + return snprintf(buf, PAGE_SIZE, "RAID-%s\n", raid_levels[ret]); +} + +static ssize_t raid_state_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct scsi_device *sdev; + struct sssraid_dev *hdev; + struct sssraid_vd_info *vd_info; + struct sssraid_sdev_hostdata *hostdata; + int ret; + + sdev = to_scsi_device(dev); + hdev = shost_priv(sdev->host); + hostdata = sdev->hostdata; + + vd_info = kmalloc(sizeof(*vd_info), GFP_KERNEL); + if (!vd_info || !SSSRAID_DEV_INFO_ATTR_VD(hostdata->attr)) + return snprintf(buf, PAGE_SIZE, "NA\n"); + + ret = sssraid_get_vd_info(hdev, vd_info, sdev->id); + if (ret) { + vd_info->vd_status = 0; + vd_info->rg_id = 0xff; + } + + ret = (vd_info->vd_status < ARRAY_SIZE(raid_states)) ? vd_info->vd_status : 0; + + kfree(vd_info); + + return snprintf(buf, PAGE_SIZE, "%s\n", raid_states[ret]); +} + +static ssize_t raid_resync_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct scsi_device *sdev; + struct sssraid_dev *hdev; + struct sssraid_vd_info *vd_info; + struct sssraid_bgtask *bgtask; + struct sssraid_sdev_hostdata *hostdata; + u8 rg_id, i, progress = 0; + int ret; + + sdev = to_scsi_device(dev); + hdev = shost_priv(sdev->host); + hostdata = sdev->hostdata; + + vd_info = kmalloc(sizeof(*vd_info), GFP_KERNEL); + if (!vd_info || !SSSRAID_DEV_INFO_ATTR_VD(hostdata->attr)) + return snprintf(buf, PAGE_SIZE, "NA\n"); + + ret = sssraid_get_vd_info(hdev, vd_info, sdev->id); + if (ret) + goto out; + + rg_id = vd_info->rg_id; + + bgtask = (struct sssraid_bgtask *)vd_info; + ret = sssraid_get_bgtask(hdev, bgtask); + if (ret) + goto out; + for (i = 0; i < bgtask->task_num; i++) { + if ((bgtask->bgtask[i].type == BGTASK_TYPE_REBUILD) && + (le16_to_cpu(bgtask->bgtask[i].vd_id) == rg_id)) + progress = bgtask->bgtask[i].progress; + } + +out: + kfree(vd_info); + return snprintf(buf, PAGE_SIZE, "%d\n", progress); +} + +static DEVICE_ATTR_RO(raid_level); +static DEVICE_ATTR_RO(raid_state); +static DEVICE_ATTR_RO(raid_resync); + +static struct device_attribute *sssraid_dev_attrs[] = { + &dev_attr_raid_level, + &dev_attr_raid_state, + &dev_attr_raid_resync, + NULL, +}; + +static struct scsi_host_template sssraid_driver_template = { + .module = THIS_MODULE, + .name = "3SNIC Logic sssraid driver", + .proc_name = "sssraid", + .queuecommand = sssraid_qcmd, + .slave_alloc = sssraid_slave_alloc, + .slave_destroy = sssraid_slave_destroy, + .slave_configure = sssraid_slave_configure, + .scan_finished = sssraid_scan_finished, + .eh_timed_out = sssraid_scmd_timeout, + .eh_abort_handler = sssraid_abort_handler, + .eh_target_reset_handler = sssraid_eh_target_reset, + .eh_bus_reset_handler = sssraid_bus_reset_handler, + .eh_host_reset_handler = sssraid_eh_host_reset, + .change_queue_depth = scsi_change_queue_depth, + .host_tagset = 0, + .this_id = -1, + .unchecked_isa_dma = 0, + .shost_attrs = sssraid_host_attrs, + .sdev_attrs = sssraid_dev_attrs, + .host_reset = sssraid_sysfs_host_reset, +}; + +/** + * sssraid_probe - PCI probe callback + * @pdev: PCI device instance + * @id: PCI device ID details + * + * controller initialization routine. + * Allocate per adapter instance through shost_priv and + * initialize controller specific data structures, initializae + * the controller hardware, add shost to the SCSI subsystem. + * + * Return: 0 on success, non-zero on failure. + */ + +static int +sssraid_probe(struct pci_dev *pdev, const struct pci_device_id *id) +{ + struct sssraid_dev *hdev; + struct Scsi_Host *shost; + int node; + char bsg_name[15]; + int retval = 0; + + node = dev_to_node(&pdev->dev); + if (node == NUMA_NO_NODE) { + node = first_memory_node; + set_dev_node(&pdev->dev, node); + } + + shost = scsi_host_alloc(&sssraid_driver_template, sizeof(*hdev)); + if (!shost) { + retval = -ENODEV; + ioc_err(hdev, "Failed to allocate scsi host\n"); + goto shost_failed; + } + + hdev = shost_priv(shost); + hdev->numa_node = node; + hdev->instance = shost->host_no; /* for device instance */ + sprintf(hdev->name, "%s%d", SSSRAID_DRIVER_NAME, hdev->instance); + + init_rwsem(&hdev->devices_rwsem); + spin_lock_init(&hdev->state_lock); + + spin_lock_init(&hdev->fwevt_lock); + spin_lock_init(&hdev->watchdog_lock); + + INIT_LIST_HEAD(&hdev->fwevt_list); + +// logging_level = 1; //garden test + hdev->logging_level = logging_level; /* according to log_debug_switch*/ + + snprintf(hdev->fwevt_worker_name, sizeof(hdev->fwevt_worker_name), + "%s%d_fwevt_wrkr", SSSRAID_DRIVER_NAME, hdev->instance); + hdev->fwevt_worker_thread = alloc_ordered_workqueue( + hdev->fwevt_worker_name, WQ_MEM_RECLAIM); + if (!hdev->fwevt_worker_thread) { + ioc_err(hdev, "failure at %s:%d/%s()!\n", + __FILE__, __LINE__, __func__); + retval = -ENODEV; + goto out_fwevtthread_failed; + } + + hdev->shost = shost; + hdev->pdev = pdev; + + if (sssraid_init_ioc(hdev, 0)) { + ioc_err(hdev, "failure at %s:%d/%s()!\n", + __FILE__, __LINE__, __func__); + retval = -ENODEV; + goto out_iocinit_failed; + } + + sssraid_shost_init(hdev); + + retval = scsi_add_host(shost, &pdev->dev); + if (retval) { + ioc_err(hdev, "failure at %s:%d/%s()!\n", + __FILE__, __LINE__, __func__); + goto addhost_failed; + } + + snprintf(bsg_name, sizeof(bsg_name), "%s%d", SSSRAID_DRIVER_NAME, shost->host_no); + hdev->bsg_queue = bsg_setup_queue(&shost->shost_gendev, bsg_name, + sssraid_bsg_host_dispatch, NULL, sssraid_cmd_size(hdev)); + if (IS_ERR(hdev->bsg_queue)) { + ioc_err(hdev, "err, setup bsg failed\n"); + hdev->bsg_queue = NULL; + goto bsg_setup_failed; + } + + sssraid_change_host_state(hdev, SSSRAID_LIVE); + + scsi_scan_host(shost); + return retval; + +bsg_setup_failed: + scsi_remove_host(shost); +addhost_failed: + sssraid_cleanup_ioc(hdev, 0); +out_iocinit_failed: + destroy_workqueue(hdev->fwevt_worker_thread); +out_fwevtthread_failed: + scsi_host_put(shost); +shost_failed: + return retval; +} + +static void sssraid_remove(struct pci_dev *pdev) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + + if (!shost) + return; + + ioc_info(hdev, "enter sssraid remove\n"); + + hdev = shost_priv(shost); + + sssraid_change_host_state(hdev, SSSRAID_DELETING); + + if (!pci_device_is_present(pdev)) + sssraid_back_all_io(hdev); + + sssraid_cleanup_fwevt_list(hdev); + destroy_workqueue(hdev->fwevt_worker_thread); + + sssraid_remove_bsg(hdev); + scsi_remove_host(shost); + sssraid_cleanup_ioc(hdev, 0); + + scsi_host_put(shost); +} + +static void sssraid_shutdown(struct pci_dev *pdev) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + + if (!shost) + return; + + hdev = shost_priv(shost); + + sssraid_cleanup_fwevt_list(hdev); + destroy_workqueue(hdev->fwevt_worker_thread); + sssraid_cleanup_ioc(hdev, 0); +} + +#ifdef CONFIG_PM +static int sssraid_suspend(struct pci_dev *pdev, pm_message_t state) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + pci_power_t device_state; + + if (!shost) + return 0; + + hdev = shost_priv(shost); + + while (hdev->state == SSSRAID_RESETTING) + ssleep(1); + sssraid_cleanup_fwevt_list(hdev); + scsi_block_requests(shost); + sssraid_cleanup_ioc(hdev, 1); + + device_state = pci_choose_state(pdev, state); + pci_save_state(pdev); + pci_set_power_state(pdev, device_state); + + return 0; +} + +static int sssraid_resume(struct pci_dev *pdev) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + pci_power_t device_state = pdev->current_state; + + if (!shost) + return 0; + + hdev = shost_priv(shost); + + ioc_info(hdev, "pdev=0x%p, slot=%s, previous operating state [D%d]\n", + pdev, pci_name(pdev), device_state); + pci_set_power_state(pdev, PCI_D0); + pci_enable_wake(pdev, PCI_D0, 0); + pci_restore_state(pdev); + hdev->pdev = pdev; + hdev->cpu_count = num_online_cpus(); + + /* sssraid_setup_resources in sssraid_init_ioc */ + sssraid_init_ioc(hdev, 1); + scsi_unblock_requests(shost); + + return 0; +} +#endif + +static pci_ers_result_t sssraid_pci_error_detected(struct pci_dev *pdev, + pci_channel_state_t state) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + + if (!shost) + return PCI_ERS_RESULT_NONE; + + hdev = shost_priv(shost); + + ioc_info(hdev, "enter pci error detect, state:%d\n", state); + + switch (state) { + case pci_channel_io_normal: + ioc_warn(hdev, "channel is normal, do nothing\n"); + + return PCI_ERS_RESULT_CAN_RECOVER; + case pci_channel_io_frozen: + ioc_warn(hdev, "channel io frozen, need reset controller\n"); + + scsi_block_requests(hdev->shost); + + sssraid_change_host_state(hdev, SSSRAID_RESETTING); + + return PCI_ERS_RESULT_NEED_RESET; + case pci_channel_io_perm_failure: + ioc_warn(hdev, "channel io failure, request disconnect\n"); + + return PCI_ERS_RESULT_DISCONNECT; + } + + return PCI_ERS_RESULT_NEED_RESET; +} + +static pci_ers_result_t sssraid_pci_slot_reset(struct pci_dev *pdev) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + + if (!shost) + return PCI_ERS_RESULT_NONE; + + hdev = shost_priv(shost); + + ioc_info(hdev, "restart after slot reset\n"); + + pci_restore_state(pdev); + + sssraid_soft_reset_handler(hdev); + + scsi_unblock_requests(hdev->shost); + + return PCI_ERS_RESULT_RECOVERED; +} + +static void sssraid_reset_done(struct pci_dev *pdev) +{ + struct Scsi_Host *shost = pci_get_drvdata(pdev); + struct sssraid_dev *hdev; + + if (!shost) + return; + + hdev = shost_priv(shost); + + ioc_info(hdev, "enter sssraid reset done\n"); +} + +static struct pci_error_handlers sssraid_err_handler = { + .error_detected = sssraid_pci_error_detected, + .slot_reset = sssraid_pci_slot_reset, + .reset_done = sssraid_reset_done, +}; + +static const struct pci_device_id sssraid_pci_id_table[] = { + { PCI_DEVICE(PCI_VENDOR_ID_3SNIC_LOGIC, SSSRAID_SERVER_DEVICE_HBA_DID) }, + { PCI_DEVICE(PCI_VENDOR_ID_3SNIC_LOGIC, SSSRAID_SERVER_DEVICE_RAID_DID) }, + { 0, } +}; +MODULE_DEVICE_TABLE(pci, sssraid_pci_id_table); + +static struct pci_driver sssraid_pci_driver = { + .name = SSSRAID_DRIVER_NAME, + .id_table = sssraid_pci_id_table, + .probe = sssraid_probe, + .remove = sssraid_remove, + .shutdown = sssraid_shutdown, +#ifdef CONFIG_PM + .suspend = sssraid_suspend, + .resume = sssraid_resume, +#endif + .err_handler = &sssraid_err_handler, +}; + +static int __init sssraid_init(void) +{ + int ret_val; + + pr_info("Loading %s version %s\n", SSSRAID_DRIVER_NAME, + SSSRAID_DRIVER_VERSION); + + sssraid_class = class_create(THIS_MODULE, "sssraid"); + if (IS_ERR(sssraid_class)) { + ret_val = PTR_ERR(sssraid_class); + return ret_val; + } + + ret_val = pci_register_driver(&sssraid_pci_driver); + + return ret_val; +} + +static void __exit sssraid_exit(void) +{ + pr_info("Unloading %s version %s\n", SSSRAID_DRIVER_NAME, + SSSRAID_DRIVER_VERSION); + + class_destroy(sssraid_class); + pci_unregister_driver(&sssraid_pci_driver); +} + +MODULE_AUTHOR("liangry1(a)3snic.com"); +MODULE_DESCRIPTION("3SNIC Information Technology SSSRAID Driver"); +MODULE_LICENSE("GPL"); +MODULE_VERSION(SSSRAID_DRIVER_VERSION); +module_init(sssraid_init); +module_exit(sssraid_exit); -- 2.25.1

2 1

[PATCH openEuler-5.10-LTS] wifi: wilc1000: validate length of IEEE80211_P2P_ATTR_CHANNEL_LIST attribute
by Zheng Zengkai 28 Dec '22

28 Dec '22

From: Phil Turnbull <philipturnbull(a)github.com> stable inclusion from stable-v5.10.157 commit 5a068535c0073c8402aa0755e8ef259fb98a33c5 category: bugfix bugzilla: 188173 https://gitee.com/src-openeuler/kernel/issues/I66M3J CVE: CVE-2022-47521 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit f9b62f9843c7b0afdaecabbcebf1dbba18599408 upstream. Validate that the IEEE80211_P2P_ATTR_CHANNEL_LIST attribute contains enough space for a 'struct wilc_attr_oper_ch'. If the attribute is too small then it can trigger an out-of-bounds write later in the function. 'struct wilc_attr_oper_ch' is variable sized so also check 'attr_len' does not extend beyond the end of 'buf'. Signed-off-by: Phil Turnbull <philipturnbull(a)github.com> Tested-by: Ajay Kathat <ajay.kathat(a)microchip.com> Acked-by: Ajay Kathat <ajay.kathat(a)microchip.com> Signed-off-by: Kalle Valo <kvalo(a)kernel.org> Link: https://lore.kernel.org/r/20221123153543.8568-4-philipturnbull@github.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Liu Jian <liujian56(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> Reviewed-by: Yue Haibing <yuehaibing(a)huawei.com> --- drivers/net/wireless/microchip/wilc1000/cfg80211.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/wireless/microchip/wilc1000/cfg80211.c b/drivers/net/wireless/microchip/wilc1000/cfg80211.c index 64e01b84d98d..dd26f2086180 100644 --- a/drivers/net/wireless/microchip/wilc1000/cfg80211.c +++ b/drivers/net/wireless/microchip/wilc1000/cfg80211.c @@ -947,7 +947,8 @@ static inline void wilc_wfi_cfg_parse_ch_attr(u8 *buf, u32 len, u8 sta_ch) if (index + sizeof(*e) + attr_size > len) return; - if (e->attr_type == IEEE80211_P2P_ATTR_CHANNEL_LIST) + if (e->attr_type == IEEE80211_P2P_ATTR_CHANNEL_LIST && + attr_size >= (sizeof(struct wilc_attr_ch_list) - sizeof(*e))) ch_list_idx = index; else if (e->attr_type == IEEE80211_P2P_ATTR_OPER_CHANNEL && attr_size == (sizeof(struct wilc_attr_oper_ch) - sizeof(*e))) -- 2.20.1

1 0

[PATCH openEuler-5.10-LTS 1/6] drm/amdkfd: Check for null pointer after calling kmemdup
by Zheng Zengkai 28 Dec '22

28 Dec '22

From: Jiasheng Jiang <jiasheng(a)iscas.ac.cn> stable inclusion from stable-v6.0.14 commit abfaf0eee97925905e742aa3b0b72e04a918fa9e category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I6694U CVE: CVE-2022-3108 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- As the possible failure of the allocation, kmemdup() may return NULL pointer. Therefore, it should be better to check the 'props2' in order to prevent the dereference of NULL pointer. Fixes: 3a87177eb141 ("drm/amdkfd: Add topology support for dGPUs") Signed-off-by: Jiasheng Jiang <jiasheng(a)iscas.ac.cn> Reviewed-by: Felix Kuehling <Felix.Kuehling(a)amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Ren Zhijie <renzhijie2(a)huawei.com> Reviewed-by: songping yu <yusongping(a)huawei.com> Reviewed-by: Chen Hui <judy.chenhui(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> --- drivers/gpu/drm/amd/amdkfd/kfd_crat.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index 86b4dadf772e..02e3c650ed1c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -408,6 +408,9 @@ static int kfd_parse_subtype_iolink(struct crat_subtype_iolink *iolink, return -ENODEV; /* same everything but the other direction */ props2 = kmemdup(props, sizeof(*props2), GFP_KERNEL); + if (!props2) + return -ENOMEM; + props2->node_from = id_to; props2->node_to = id_from; props2->kobj = NULL; -- 2.20.1

1 5

[PATCH openEuler-1.0-LTS] dm thin: Use last transaction's pmd->root when commit failed
by Yongqiang Liu 27 Dec '22

27 Dec '22

From: Zhihao Cheng <chengzhihao1(a)huawei.com> mainline inclusion from mainline-v6.2-rc1 commit 7991dbff6849f67e823b7cc0c15e5a90b0549b9f category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I65M32 CVE: NA -------------------------------- Recently we found a softlock up problem in dm thin pool btree lookup code due to corrupted metadata: Kernel panic - not syncing: softlockup: hung tasks CPU: 7 PID: 2669225 Comm: kworker/u16:3 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) Workqueue: dm-thin do_worker [dm_thin_pool] Call Trace: <IRQ> dump_stack+0x9c/0xd3 panic+0x35d/0x6b9 watchdog_timer_fn.cold+0x16/0x25 __run_hrtimer+0xa2/0x2d0 </IRQ> RIP: 0010:__relink_lru+0x102/0x220 [dm_bufio] __bufio_new+0x11f/0x4f0 [dm_bufio] new_read+0xa3/0x1e0 [dm_bufio] dm_bm_read_lock+0x33/0xd0 [dm_persistent_data] ro_step+0x63/0x100 [dm_persistent_data] btree_lookup_raw.constprop.0+0x44/0x220 [dm_persistent_data] dm_btree_lookup+0x16f/0x210 [dm_persistent_data] dm_thin_find_block+0x12c/0x210 [dm_thin_pool] __process_bio_read_only+0xc5/0x400 [dm_thin_pool] process_thin_deferred_bios+0x1a4/0x4a0 [dm_thin_pool] process_one_work+0x3c5/0x730 Following process may generate a broken btree mixed with fresh and stale btree nodes, which could get dm thin trapped in an infinite loop while looking up data block: Transaction 1: pmd->root = A, A->B->C // One path in btree pmd->root = X, X->Y->Z // Copy-up Transaction 2: X,Z is updated on disk, Y write failed. // Commit failed, dm thin becomes read-only. process_bio_read_only dm_thin_find_block __find_block dm_btree_lookup(pmd->root) The pmd->root points to a broken btree, Y may contain stale node pointing to any block, for example X, which gets dm thin trapped into a dead loop while looking up Z. Fix this by setting pmd->root in __open_metadata(), so that dm thin will use the last transaction's pmd->root if commit failed. Fetch a reproducer in [Link]. Linke: https://bugzilla.kernel.org/show_bug.cgi?id=216790 Cc: stable(a)vger.kernel.org Fixes: 991d9fa02da0 ("dm: add thin provisioning target") Signed-off-by: Zhihao Cheng <chengzhihao1(a)huawei.com> Acked-by: Joe Thornber <ejt(a)redhat.com> Signed-off-by: Mike Snitzer <snitzer(a)kernel.org> Signed-off-by: Zhihao Cheng <chengzhihao1(a)huawei.com> Reviewed-by: Zhang Yi <yi.zhang(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- drivers/md/dm-thin-metadata.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/md/dm-thin-metadata.c b/drivers/md/dm-thin-metadata.c index 80c4f27e99f9..26f04b84e8a2 100644 --- a/drivers/md/dm-thin-metadata.c +++ b/drivers/md/dm-thin-metadata.c @@ -660,6 +660,15 @@ static int __open_metadata(struct dm_pool_metadata *pmd) goto bad_cleanup_data_sm; } + /* + * For pool metadata opening process, root setting is redundant + * because it will be set again in __begin_transaction(). But dm + * pool aborting process really needs to get last transaction's + * root to avoid accessing broken btree. + */ + pmd->root = le64_to_cpu(disk_super->data_mapping_root); + pmd->details_root = le64_to_cpu(disk_super->device_details_root); + __setup_btree_details(pmd); dm_bm_unlock(sblock); -- 2.25.1

1 0

[PATCH openEuler-5.10] timekeeping: Adding a padding before timekeeper in tk_core
by Zheng Zengkai 26 Dec '22

26 Dec '22

From: Jialin Zhang <zhangjialin11(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I47W8L CVE: NA --------------------------- Adding a padding before timekeeper in tk_core can futhor fix the false sharing problem discovered by commit 605276666775 ("timekeeping: Avoiding false sharing in field access of tk_core") context1 of unixbench is tested on Huawei Kunpeng920. before this patch 0 CPUs in system; running 96 parallel copies of tests Pipe-based Context Switching 17213571.7 lps (10.0 s, 1 samples) System Benchmarks Partial Index BASELINE RESULT INDEX Pipe-based Context Switching 4000.0 17213571.7 43033.9 ======== System Benchmarks Index Score (Partial Only) 43033.9 after this patch 0 CPUs in system; running 96 parallel copies of tests Pipe-based Context Switching 20321268.3 lps (10.0 s, 1 samples) System Benchmarks Partial Index BASELINE RESULT INDEX Pipe-based Context Switching 4000.0 20321268.3 50803.2 ======== System Benchmarks Index Score (Partial Only) 50803.2 Signed-off-by: Jialin Zhang <zhangjialin11(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> --- kernel/time/timekeeping.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index e3142bf7648e..c8d9b89a7571 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -52,6 +52,9 @@ static struct { u64 padding[8]; #endif seqcount_raw_spinlock_t seq; +#ifdef CONFIG_ARCH_LLC_128_LINE_SIZE + u64 padding2[2]; +#endif struct timekeeper timekeeper; #ifdef CONFIG_ARCH_LLC_128_LINE_SIZE } tk_core ____cacheline_aligned_128 = { -- 2.20.1

1 0

[PATCH openEuler-5.10] timekeeping: Adding a padding before timekeeper in tk_core
by Zheng Zengkai 26 Dec '22

26 Dec '22

From: Jialin Zhang <zhangjialin11(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I47W8L CVE: NA --------------------------- Adding a padding before timekeeper in tk_core can futhor fix the false sharing problem discovered by commit 605276666775 ("timekeeping: Avoiding false sharing in field access of tk_core") context1 of unixbench is tested on Huawei Kunpeng920. before this patch 0 CPUs in system; running 96 parallel copies of tests Pipe-based Context Switching 17213571.7 lps (10.0 s, 1 samples) System Benchmarks Partial Index BASELINE RESULT INDEX Pipe-based Context Switching 4000.0 17213571.7 43033.9 ======== System Benchmarks Index Score (Partial Only) 43033.9 after this patch 0 CPUs in system; running 96 parallel copies of tests Pipe-based Context Switching 20321268.3 lps (10.0 s, 1 samples) System Benchmarks Partial Index BASELINE RESULT INDEX Pipe-based Context Switching 4000.0 20321268.3 50803.2 ======== System Benchmarks Index Score (Partial Only) 50803.2 Signed-off-by: Jialin Zhang <zhangjialin11(a)huawei.com> Reviewed-by: Xie XiuQi <xiexiuqi(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> --- kernel/time/timekeeping.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index e3142bf7648e..c8d9b89a7571 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -52,6 +52,9 @@ static struct { u64 padding[8]; #endif seqcount_raw_spinlock_t seq; +#ifdef CONFIG_ARCH_LLC_128_LINE_SIZE + u64 padding2[2]; +#endif struct timekeeper timekeeper; #ifdef CONFIG_ARCH_LLC_128_LINE_SIZE } tk_core ____cacheline_aligned_128 = { -- 2.20.1

1 0

[PATCH openEuler-1.0-LTS 1/2] power: supply: wm8350-power: Add missing free in free_charger_irq
by Yongqiang Liu 24 Dec '22

24 Dec '22

From: Jiasheng Jiang <jiasheng(a)iscas.ac.cn> stable inclusion from stable-v4.19.269 commit 60dd1082322966f192f42fe2a6605dfa08eef41f category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I6694F CVE: CVE-2022-3111 -------------------------------- [ Upstream commit 6dee930f6f6776d1e5a7edf542c6863b47d9f078 ] In free_charger_irq(), there is no free for 'WM8350_IRQ_CHG_FAST_RDY'. Therefore, it should be better to add it in order to avoid the memory leak. Fixes: 14431aa0c5a4 ("power_supply: Add support for WM8350 PMU") Signed-off-by: Jiasheng Jiang <jiasheng(a)iscas.ac.cn> Acked-by: Charles Keepax <ckeepax(a)opensource.cirrus.com> Signed-off-by: Sebastian Reichel <sebastian.reichel(a)collabora.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Chen jun <chenjun102(a)huawei.com> Reviewed-by: chenweilong <chenweilong(a)huawei.com> Reviewed-by: Wang Weiyang <wangweiyang2(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- drivers/power/supply/wm8350_power.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/power/supply/wm8350_power.c b/drivers/power/supply/wm8350_power.c index 15c0ca15e2aa..034ddfe3b62c 100644 --- a/drivers/power/supply/wm8350_power.c +++ b/drivers/power/supply/wm8350_power.c @@ -459,6 +459,7 @@ static void free_charger_irq(struct wm8350 *wm8350) wm8350_free_irq(wm8350, WM8350_IRQ_CHG_TO, wm8350); wm8350_free_irq(wm8350, WM8350_IRQ_CHG_END, wm8350); wm8350_free_irq(wm8350, WM8350_IRQ_CHG_START, wm8350); + wm8350_free_irq(wm8350, WM8350_IRQ_CHG_FAST_RDY, wm8350); wm8350_free_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_3P9, wm8350); wm8350_free_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_3P1, wm8350); wm8350_free_irq(wm8350, WM8350_IRQ_CHG_VBATT_LT_2P85, wm8350); -- 2.25.1

1 1

[PATCH openEuler-1.0-LTS] media: dvb-core: Fix UAF due to refcount races at releasing
by Yongqiang Liu 24 Dec '22

24 Dec '22

From: Takashi Iwai <tiwai(a)suse.de> stable inclusion from stable-v6.1.1 commit fd3d91ab1c6ab0628fe642dd570b56302c30a792 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I5SDE4 CVE: CVE-2022-41218 -------------------------------- The dvb-core tries to sync the releases of opened files at dvb_dmxdev_release() with two refcounts: dvbdev->users and dvr_dvbdev->users. A problem is present in those two syncs: when yet another dvb_demux_open() is called during those sync waits, dvb_demux_open() continues to process even if the device is being closed. This includes the increment of the former refcount, resulting in the leftover refcount after the sync of the latter refcount at dvb_dmxdev_release(). It ends up with use-after-free, since the function believes that all usages were gone and releases the resources. This patch addresses the problem by adding the check of dmxdev->exit flag at dvb_demux_open(), just like dvb_dvr_open() already does. With the exit flag check, the second call of dvb_demux_open() fails, hence the further corruption can be avoided. Also for avoiding the races of the dmxdev->exit flag reference, this patch serializes the dmxdev->exit set up and the sync waits with the dmxdev->mutex lock at dvb_dmxdev_release(). Without the mutex lock, dvb_demux_open() (or dvb_dvr_open()) may run concurrently with dvb_dmxdev_release(), which allows to skip the exit flag check and continue the open process that is being closed. CVE-2022-41218 is assigned to those bugs above. Reported-by: Hyunwoo Kim <imv4bel(a)gmail.com> Cc: <stable(a)vger.kernel.org> Link: https://lore.kernel.org/20220908132754.30532-1-tiwai@suse.de Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Signed-off-by: Hans Verkuil <hverkuil-cisco(a)xs4all.nl> Signed-off-by: Ren Zhijie <renzhijie2(a)huawei.com> Reviewed-by: songping yu <yusongping(a)huawei.com> Reviewed-by: Zhang Qiao <zhangqiao22(a)huawei.com> Reviewed-by: Xiu Jianfeng <xiujianfeng(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- drivers/media/dvb-core/dmxdev.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/media/dvb-core/dmxdev.c b/drivers/media/dvb-core/dmxdev.c index d548f98c7a67..6f0121aa546e 100644 --- a/drivers/media/dvb-core/dmxdev.c +++ b/drivers/media/dvb-core/dmxdev.c @@ -800,6 +800,11 @@ static int dvb_demux_open(struct inode *inode, struct file *file) if (mutex_lock_interruptible(&dmxdev->mutex)) return -ERESTARTSYS; + if (dmxdev->exit) { + mutex_unlock(&dmxdev->mutex); + return -ENODEV; + } + for (i = 0; i < dmxdev->filternum; i++) if (dmxdev->filter[i].state == DMXDEV_STATE_FREE) break; @@ -1445,7 +1450,10 @@ EXPORT_SYMBOL(dvb_dmxdev_init); void dvb_dmxdev_release(struct dmxdev *dmxdev) { + mutex_lock(&dmxdev->mutex); dmxdev->exit = 1; + mutex_unlock(&dmxdev->mutex); + if (dmxdev->dvbdev->users > 1) { wait_event(dmxdev->dvbdev->wait_queue, dmxdev->dvbdev->users == 1); -- 2.25.1

1 0

[PATCH openEuler-1.0-LTS] sched: Reinit task's vruntime if a task sleep over 200 days
by Yongqiang Liu 24 Dec '22

24 Dec '22

From: Zhang Qiao <zhangqiao22(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I67BL1 CVE: NA ------------------------------- If a task sleep for long time, it maybe cause a s64 overflow issue at max_vruntime() and the task will be set an incorrect vruntime, lead to the task be starve. For fix it, we set the task's vruntime as cfs_rq->min_vruntime when wakeup. Signed-off-by: Zhang Qiao <zhangqiao22(a)huawei.com> Reviewed-by: Chen Hui <judy.chenhui(a)huawei.com> Reviewed-by: songping yu <yusongping(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- kernel/sched/fair.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 0dba06ce0677..ad6a7923c9ed 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3949,10 +3949,14 @@ static void check_spread(struct cfs_rq *cfs_rq, struct sched_entity *se) #endif } +/* The threshold when the wakee's vruntime should be set cfs_rq->min_vruntime, default: 200 days */ +#define WAKEUP_REINIT_THRESHOLD_NS (200 * 24 * 3600 * NSEC_PER_SEC) + static void place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) { u64 vruntime = cfs_rq->min_vruntime; + struct rq *rq = rq_of(cfs_rq); /* * The 'current' period is already promised to the current tasks, @@ -3977,8 +3981,11 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) vruntime -= thresh; } + if (unlikely(!initial && (s64)(rq_clock_task(rq) - se->exec_start) > WAKEUP_REINIT_THRESHOLD_NS)) + se->vruntime = vruntime; + else /* ensure we never gain time by being placed backwards. */ - se->vruntime = max_vruntime(se->vruntime, vruntime); + se->vruntime = max_vruntime(se->vruntime, vruntime); } static void check_enqueue_throttle(struct cfs_rq *cfs_rq); -- 2.25.1

1 0

[PATCH OLK-5.10 1/2] Revert "scsi: iscsi: fix kabi broken in struct iscsi_transport"
by Yongqiang Liu 23 Dec '22

23 Dec '22

From: Li Nan <linan122(a)huawei.com> hulk inclusion category: bugfix bugzilla: 188176, https://gitee.com/openeuler/kernel/issues/I67294 CVE: NA -------------------------------- This reverts commit 230035efb98b6edc066f3671e092a25013a131a2. Drivers use tgt_dscvr will compile failed because API has changed. Signed-off-by: Li Nan <linan122(a)huawei.com> Reviewed-by: Zhang Yi <yi.zhang(a)huawei.com> Reviewed-by: Zhang Yi <yi.zhang(a)huawei.com> --- drivers/infiniband/ulp/iser/iscsi_iser.c | 9 ++----- drivers/scsi/be2iscsi/be_main.c | 9 ++----- drivers/scsi/bnx2i/bnx2i_iscsi.c | 9 ++----- drivers/scsi/cxgbi/cxgb3i/cxgb3i.c | 9 ++----- drivers/scsi/cxgbi/cxgb4i/cxgb4i.c | 9 ++----- drivers/scsi/qedi/qedi_iscsi.c | 8 ++---- drivers/scsi/qla4xxx/ql4_os.c | 8 ++---- drivers/scsi/scsi_transport_iscsi.c | 31 ++++++------------------ include/scsi/iscsi_if.h | 1 - include/scsi/scsi_transport_iscsi.h | 17 +------------ 10 files changed, 23 insertions(+), 87 deletions(-) diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index 4884b122e413..a16e066989fa 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -979,22 +979,17 @@ static struct scsi_host_template iscsi_iser_sht = { .track_queue_depth = 1, }; -static struct iscsi_transport_expand iscsi_iser_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - static struct iscsi_transport iscsi_iser_transport = { .owner = THIS_MODULE, .name = "iser", - .caps = CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_TEXT_NEGO - | CAP_OPS_EXPAND, + .caps = CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_TEXT_NEGO, /* session management */ .create_session = iscsi_iser_session_create, .destroy_session = iscsi_iser_session_destroy, /* connection management */ .create_conn = iscsi_iser_conn_create, .bind_conn = iscsi_iser_conn_bind, - .ops_expand = &iscsi_iser_expand, + .unbind_conn = iscsi_conn_unbind, .destroy_conn = iscsi_conn_teardown, .attr_is_visible = iser_attr_is_visible, .set_param = iscsi_iser_set_param, diff --git a/drivers/scsi/be2iscsi/be_main.c b/drivers/scsi/be2iscsi/be_main.c index 06f697bfc49f..b977e039bb78 100644 --- a/drivers/scsi/be2iscsi/be_main.c +++ b/drivers/scsi/be2iscsi/be_main.c @@ -5801,21 +5801,16 @@ static struct pci_error_handlers beiscsi_eeh_handlers = { .resume = beiscsi_eeh_resume, }; -struct iscsi_transport_expand beiscsi_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - struct iscsi_transport beiscsi_iscsi_transport = { .owner = THIS_MODULE, .name = DRV_NAME, .caps = CAP_RECOVERY_L0 | CAP_HDRDGST | CAP_TEXT_NEGO | - CAP_MULTI_R2T | CAP_DATADGST | CAP_DATA_PATH_OFFLOAD | - CAP_OPS_EXPAND, + CAP_MULTI_R2T | CAP_DATADGST | CAP_DATA_PATH_OFFLOAD, .create_session = beiscsi_session_create, .destroy_session = beiscsi_session_destroy, .create_conn = beiscsi_conn_create, .bind_conn = beiscsi_conn_bind, - .ops_expand = &beiscsi_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .destroy_conn = iscsi_conn_teardown, .attr_is_visible = beiscsi_attr_is_visible, .set_iface_param = beiscsi_iface_set_param, diff --git a/drivers/scsi/bnx2i/bnx2i_iscsi.c b/drivers/scsi/bnx2i/bnx2i_iscsi.c index e13c77a76150..8cf2f9a7cfdc 100644 --- a/drivers/scsi/bnx2i/bnx2i_iscsi.c +++ b/drivers/scsi/bnx2i/bnx2i_iscsi.c @@ -2274,23 +2274,18 @@ static struct scsi_host_template bnx2i_host_template = { .track_queue_depth = 1, }; - -static struct iscsi_transport_expand bnx2i_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - struct iscsi_transport bnx2i_iscsi_transport = { .owner = THIS_MODULE, .name = "bnx2i", .caps = CAP_RECOVERY_L0 | CAP_HDRDGST | CAP_MULTI_R2T | CAP_DATADGST | CAP_DATA_PATH_OFFLOAD | - CAP_TEXT_NEGO | CAP_OPS_EXPAND, + CAP_TEXT_NEGO, .create_session = bnx2i_session_create, .destroy_session = bnx2i_session_destroy, .create_conn = bnx2i_conn_create, .bind_conn = bnx2i_conn_bind, - .ops_expand = &bnx2i_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .destroy_conn = bnx2i_conn_destroy, .attr_is_visible = bnx2i_attr_is_visible, .set_param = iscsi_set_param, diff --git a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c index 65eb6230d390..edcd3fab6973 100644 --- a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c +++ b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c @@ -100,18 +100,13 @@ static struct scsi_host_template cxgb3i_host_template = { .track_queue_depth = 1, }; -static struct iscsi_transport_expand cxgb3i_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - static struct iscsi_transport cxgb3i_iscsi_transport = { .owner = THIS_MODULE, .name = DRV_MODULE_NAME, /* owner and name should be set already */ .caps = CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_HDRDGST | CAP_DATADGST | CAP_DIGEST_OFFLOAD | - CAP_PADDING_OFFLOAD | CAP_TEXT_NEGO | - CAP_OPS_EXPAND, + CAP_PADDING_OFFLOAD | CAP_TEXT_NEGO, .attr_is_visible = cxgbi_attr_is_visible, .get_host_param = cxgbi_get_host_param, .set_host_param = cxgbi_set_host_param, @@ -122,7 +117,7 @@ static struct iscsi_transport cxgb3i_iscsi_transport = { /* connection management */ .create_conn = cxgbi_create_conn, .bind_conn = cxgbi_bind_conn, - .ops_expand = &cxgb3i_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .destroy_conn = iscsi_tcp_conn_teardown, .start_conn = iscsi_conn_start, .stop_conn = iscsi_conn_stop, diff --git a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c index 88f96964c356..efb3e2b3398e 100644 --- a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c +++ b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c @@ -118,17 +118,12 @@ static struct scsi_host_template cxgb4i_host_template = { .track_queue_depth = 1, }; -static struct iscsi_transport_expand cxgb4i_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - static struct iscsi_transport cxgb4i_iscsi_transport = { .owner = THIS_MODULE, .name = DRV_MODULE_NAME, .caps = CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_HDRDGST | CAP_DATADGST | CAP_DIGEST_OFFLOAD | - CAP_PADDING_OFFLOAD | CAP_TEXT_NEGO | - CAP_OPS_EXPAND, + CAP_PADDING_OFFLOAD | CAP_TEXT_NEGO, .attr_is_visible = cxgbi_attr_is_visible, .get_host_param = cxgbi_get_host_param, .set_host_param = cxgbi_set_host_param, @@ -139,7 +134,7 @@ static struct iscsi_transport cxgb4i_iscsi_transport = { /* connection management */ .create_conn = cxgbi_create_conn, .bind_conn = cxgbi_bind_conn, - .ops_expand = &cxgb4i_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .destroy_conn = iscsi_tcp_conn_teardown, .start_conn = iscsi_conn_start, .stop_conn = iscsi_conn_stop, diff --git a/drivers/scsi/qedi/qedi_iscsi.c b/drivers/scsi/qedi/qedi_iscsi.c index 8003b3519b95..3bcadb3dd40d 100644 --- a/drivers/scsi/qedi/qedi_iscsi.c +++ b/drivers/scsi/qedi/qedi_iscsi.c @@ -1429,20 +1429,16 @@ static void qedi_cleanup_task(struct iscsi_task *task) cmd->scsi_cmd = NULL; } -static struct iscsi_transport_expand qedi_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - struct iscsi_transport qedi_iscsi_transport = { .owner = THIS_MODULE, .name = QEDI_MODULE_NAME, .caps = CAP_RECOVERY_L0 | CAP_HDRDGST | CAP_MULTI_R2T | CAP_DATADGST | - CAP_DATA_PATH_OFFLOAD | CAP_TEXT_NEGO | CAP_OPS_EXPAND, + CAP_DATA_PATH_OFFLOAD | CAP_TEXT_NEGO, .create_session = qedi_session_create, .destroy_session = qedi_session_destroy, .create_conn = qedi_conn_create, .bind_conn = qedi_conn_bind, - .ops_expand = &qedi_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .start_conn = qedi_conn_start, .stop_conn = iscsi_conn_stop, .destroy_conn = qedi_conn_destroy, diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 377d83762099..8d82d2a83059 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -246,24 +246,20 @@ static struct scsi_host_template qla4xxx_driver_template = { .vendor_id = SCSI_NL_VID_TYPE_PCI | PCI_VENDOR_ID_QLOGIC, }; -static struct iscsi_transport_expand qla4xxx_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - static struct iscsi_transport qla4xxx_iscsi_transport = { .owner = THIS_MODULE, .name = DRIVER_NAME, .caps = CAP_TEXT_NEGO | CAP_DATA_PATH_OFFLOAD | CAP_HDRDGST | CAP_DATADGST | CAP_LOGIN_OFFLOAD | - CAP_MULTI_R2T | CAP_OPS_EXPAND, + CAP_MULTI_R2T, .attr_is_visible = qla4_attr_is_visible, .create_session = qla4xxx_session_create, .destroy_session = qla4xxx_session_destroy, .start_conn = qla4xxx_conn_start, .create_conn = qla4xxx_conn_create, .bind_conn = qla4xxx_conn_bind, - .ops_expand = &qla4xxx_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .stop_conn = iscsi_conn_stop, .destroy_conn = qla4xxx_conn_destroy, .set_param = iscsi_set_param, diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index a213362524c9..517a3c4d879b 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -2257,11 +2257,7 @@ static void iscsi_ep_disconnect(struct iscsi_cls_conn *conn, bool is_active) ep = conn->ep; conn->ep = NULL; - if (session->transport->caps & CAP_OPS_EXPAND && - session->transport->ops_expand && - session->transport->ops_expand->unbind_conn) - session->transport->ops_expand->unbind_conn(conn, is_active); - + session->transport->unbind_conn(conn, is_active); session->transport->ep_disconnect(ep); ISCSI_DBG_TRANS_CONN(conn, "disconnect ep done.\n"); } @@ -3228,19 +3224,10 @@ iscsi_tgt_dscvr(struct iscsi_transport *transport, struct Scsi_Host *shost; struct sockaddr *dst_addr; int err; - int (*tgt_dscvr)(struct Scsi_Host *shost, enum iscsi_tgt_dscvr type, - uint32_t enable, struct sockaddr *dst_addr); - if (transport->caps & CAP_OPS_EXPAND) { - if (!transport->ops_expand || !transport->ops_expand->tgt_dscvr) - return -EINVAL; - tgt_dscvr = transport->ops_expand->tgt_dscvr; - } else { - if (!transport->ops_expand) - return -EINVAL; - tgt_dscvr = (int (*)(struct Scsi_Host *, enum iscsi_tgt_dscvr, uint32_t, - struct sockaddr *))(transport->ops_expand); - } + if (!transport->tgt_dscvr) + return -EINVAL; + shost = scsi_host_lookup(ev->u.tgt_dscvr.host_no); if (!shost) { printk(KERN_ERR "target discovery could not find host no %u\n", @@ -3250,8 +3237,8 @@ iscsi_tgt_dscvr(struct iscsi_transport *transport, dst_addr = (struct sockaddr *)((char*)ev + sizeof(*ev)); - err = tgt_dscvr(shost, ev->u.tgt_dscvr.type, - ev->u.tgt_dscvr.enable, dst_addr); + err = transport->tgt_dscvr(shost, ev->u.tgt_dscvr.type, + ev->u.tgt_dscvr.enable, dst_addr); scsi_host_put(shost); return err; } @@ -4904,10 +4891,8 @@ iscsi_register_transport(struct iscsi_transport *tt) int err; BUG_ON(!tt); - if (tt->caps & CAP_OPS_EXPAND) { - BUG_ON(!tt->ops_expand); - WARN_ON(tt->ep_disconnect && !tt->ops_expand->unbind_conn); - } + WARN_ON(tt->ep_disconnect && !tt->unbind_conn); + priv = iscsi_if_transport_lookup(tt); if (priv) return NULL; diff --git a/include/scsi/iscsi_if.h b/include/scsi/iscsi_if.h index 5679b9fb2b1e..5225a23f2d0e 100644 --- a/include/scsi/iscsi_if.h +++ b/include/scsi/iscsi_if.h @@ -761,7 +761,6 @@ enum iscsi_ping_status_code { and verification */ #define CAP_LOGIN_OFFLOAD 0x4000 /* offload session login */ -#define CAP_OPS_EXPAND 0x8000 /* oiscsi_transport->ops_expand flag */ /* * These flags describes reason of stop_conn() call */ diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h index 1f7574d89822..b7fd8b738aa8 100644 --- a/include/scsi/scsi_transport_iscsi.h +++ b/include/scsi/scsi_transport_iscsi.h @@ -29,15 +29,6 @@ struct bsg_job; struct iscsi_bus_flash_session; struct iscsi_bus_flash_conn; -/* - * The expansion of iscsi_transport to fix kabi while adding members. - */ -struct iscsi_transport_expand { - int (*tgt_dscvr)(struct Scsi_Host *shost, enum iscsi_tgt_dscvr type, - uint32_t enable, struct sockaddr *dst_addr); - void (*unbind_conn)(struct iscsi_cls_conn *conn, bool is_active); -}; - /** * struct iscsi_transport - iSCSI Transport template * @@ -91,6 +82,7 @@ struct iscsi_transport { void (*destroy_session) (struct iscsi_cls_session *session); struct iscsi_cls_conn *(*create_conn) (struct iscsi_cls_session *sess, uint32_t cid); + void (*unbind_conn) (struct iscsi_cls_conn *conn, bool is_active); int (*bind_conn) (struct iscsi_cls_session *session, struct iscsi_cls_conn *cls_conn, uint64_t transport_eph, int is_leading); @@ -132,15 +124,8 @@ struct iscsi_transport { int non_blocking); int (*ep_poll) (struct iscsi_endpoint *ep, int timeout_ms); void (*ep_disconnect) (struct iscsi_endpoint *ep); -#ifdef __GENKSYMS__ int (*tgt_dscvr) (struct Scsi_Host *shost, enum iscsi_tgt_dscvr type, uint32_t enable, struct sockaddr *dst_addr); -#else - /* - * onece ops_expand is used, caps must be set to CAP_OPS_EXPAND - */ - struct iscsi_transport_expand *ops_expand; -#endif int (*set_path) (struct Scsi_Host *shost, struct iscsi_path *params); int (*set_iface_param) (struct Scsi_Host *shost, void *data, uint32_t len); -- 2.27.0

1 1

[PATCH openEuler-5.10 1/2] Revert "scsi: iscsi: fix kabi broken in struct iscsi_transport"
by Zheng Zengkai 22 Dec '22

22 Dec '22

From: Li Nan <linan122(a)huawei.com> hulk inclusion category: bugfix bugzilla: 188176, https://gitee.com/openeuler/kernel/issues/I67294 CVE: NA -------------------------------- This reverts commit 230035efb98b6edc066f3671e092a25013a131a2. Drivers use tgt_dscvr will compile failed because API has changed. Signed-off-by: Li Nan <linan122(a)huawei.com> Reviewed-by: Zhang Yi <yi.zhang(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> --- drivers/infiniband/ulp/iser/iscsi_iser.c | 9 ++----- drivers/scsi/be2iscsi/be_main.c | 9 ++----- drivers/scsi/bnx2i/bnx2i_iscsi.c | 9 ++----- drivers/scsi/cxgbi/cxgb3i/cxgb3i.c | 9 ++----- drivers/scsi/cxgbi/cxgb4i/cxgb4i.c | 9 ++----- drivers/scsi/qedi/qedi_iscsi.c | 8 ++---- drivers/scsi/qla4xxx/ql4_os.c | 8 ++---- drivers/scsi/scsi_transport_iscsi.c | 31 ++++++------------------ include/scsi/iscsi_if.h | 1 - include/scsi/scsi_transport_iscsi.h | 17 +------------ 10 files changed, 23 insertions(+), 87 deletions(-) diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index 4884b122e413..a16e066989fa 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -979,22 +979,17 @@ static struct scsi_host_template iscsi_iser_sht = { .track_queue_depth = 1, }; -static struct iscsi_transport_expand iscsi_iser_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - static struct iscsi_transport iscsi_iser_transport = { .owner = THIS_MODULE, .name = "iser", - .caps = CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_TEXT_NEGO - | CAP_OPS_EXPAND, + .caps = CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_TEXT_NEGO, /* session management */ .create_session = iscsi_iser_session_create, .destroy_session = iscsi_iser_session_destroy, /* connection management */ .create_conn = iscsi_iser_conn_create, .bind_conn = iscsi_iser_conn_bind, - .ops_expand = &iscsi_iser_expand, + .unbind_conn = iscsi_conn_unbind, .destroy_conn = iscsi_conn_teardown, .attr_is_visible = iser_attr_is_visible, .set_param = iscsi_iser_set_param, diff --git a/drivers/scsi/be2iscsi/be_main.c b/drivers/scsi/be2iscsi/be_main.c index 06f697bfc49f..b977e039bb78 100644 --- a/drivers/scsi/be2iscsi/be_main.c +++ b/drivers/scsi/be2iscsi/be_main.c @@ -5801,21 +5801,16 @@ static struct pci_error_handlers beiscsi_eeh_handlers = { .resume = beiscsi_eeh_resume, }; -struct iscsi_transport_expand beiscsi_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - struct iscsi_transport beiscsi_iscsi_transport = { .owner = THIS_MODULE, .name = DRV_NAME, .caps = CAP_RECOVERY_L0 | CAP_HDRDGST | CAP_TEXT_NEGO | - CAP_MULTI_R2T | CAP_DATADGST | CAP_DATA_PATH_OFFLOAD | - CAP_OPS_EXPAND, + CAP_MULTI_R2T | CAP_DATADGST | CAP_DATA_PATH_OFFLOAD, .create_session = beiscsi_session_create, .destroy_session = beiscsi_session_destroy, .create_conn = beiscsi_conn_create, .bind_conn = beiscsi_conn_bind, - .ops_expand = &beiscsi_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .destroy_conn = iscsi_conn_teardown, .attr_is_visible = beiscsi_attr_is_visible, .set_iface_param = beiscsi_iface_set_param, diff --git a/drivers/scsi/bnx2i/bnx2i_iscsi.c b/drivers/scsi/bnx2i/bnx2i_iscsi.c index e13c77a76150..8cf2f9a7cfdc 100644 --- a/drivers/scsi/bnx2i/bnx2i_iscsi.c +++ b/drivers/scsi/bnx2i/bnx2i_iscsi.c @@ -2274,23 +2274,18 @@ static struct scsi_host_template bnx2i_host_template = { .track_queue_depth = 1, }; - -static struct iscsi_transport_expand bnx2i_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - struct iscsi_transport bnx2i_iscsi_transport = { .owner = THIS_MODULE, .name = "bnx2i", .caps = CAP_RECOVERY_L0 | CAP_HDRDGST | CAP_MULTI_R2T | CAP_DATADGST | CAP_DATA_PATH_OFFLOAD | - CAP_TEXT_NEGO | CAP_OPS_EXPAND, + CAP_TEXT_NEGO, .create_session = bnx2i_session_create, .destroy_session = bnx2i_session_destroy, .create_conn = bnx2i_conn_create, .bind_conn = bnx2i_conn_bind, - .ops_expand = &bnx2i_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .destroy_conn = bnx2i_conn_destroy, .attr_is_visible = bnx2i_attr_is_visible, .set_param = iscsi_set_param, diff --git a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c index 65eb6230d390..edcd3fab6973 100644 --- a/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c +++ b/drivers/scsi/cxgbi/cxgb3i/cxgb3i.c @@ -100,18 +100,13 @@ static struct scsi_host_template cxgb3i_host_template = { .track_queue_depth = 1, }; -static struct iscsi_transport_expand cxgb3i_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - static struct iscsi_transport cxgb3i_iscsi_transport = { .owner = THIS_MODULE, .name = DRV_MODULE_NAME, /* owner and name should be set already */ .caps = CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_HDRDGST | CAP_DATADGST | CAP_DIGEST_OFFLOAD | - CAP_PADDING_OFFLOAD | CAP_TEXT_NEGO | - CAP_OPS_EXPAND, + CAP_PADDING_OFFLOAD | CAP_TEXT_NEGO, .attr_is_visible = cxgbi_attr_is_visible, .get_host_param = cxgbi_get_host_param, .set_host_param = cxgbi_set_host_param, @@ -122,7 +117,7 @@ static struct iscsi_transport cxgb3i_iscsi_transport = { /* connection management */ .create_conn = cxgbi_create_conn, .bind_conn = cxgbi_bind_conn, - .ops_expand = &cxgb3i_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .destroy_conn = iscsi_tcp_conn_teardown, .start_conn = iscsi_conn_start, .stop_conn = iscsi_conn_stop, diff --git a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c index 88f96964c356..efb3e2b3398e 100644 --- a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c +++ b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c @@ -118,17 +118,12 @@ static struct scsi_host_template cxgb4i_host_template = { .track_queue_depth = 1, }; -static struct iscsi_transport_expand cxgb4i_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - static struct iscsi_transport cxgb4i_iscsi_transport = { .owner = THIS_MODULE, .name = DRV_MODULE_NAME, .caps = CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_HDRDGST | CAP_DATADGST | CAP_DIGEST_OFFLOAD | - CAP_PADDING_OFFLOAD | CAP_TEXT_NEGO | - CAP_OPS_EXPAND, + CAP_PADDING_OFFLOAD | CAP_TEXT_NEGO, .attr_is_visible = cxgbi_attr_is_visible, .get_host_param = cxgbi_get_host_param, .set_host_param = cxgbi_set_host_param, @@ -139,7 +134,7 @@ static struct iscsi_transport cxgb4i_iscsi_transport = { /* connection management */ .create_conn = cxgbi_create_conn, .bind_conn = cxgbi_bind_conn, - .ops_expand = &cxgb4i_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .destroy_conn = iscsi_tcp_conn_teardown, .start_conn = iscsi_conn_start, .stop_conn = iscsi_conn_stop, diff --git a/drivers/scsi/qedi/qedi_iscsi.c b/drivers/scsi/qedi/qedi_iscsi.c index 8003b3519b95..3bcadb3dd40d 100644 --- a/drivers/scsi/qedi/qedi_iscsi.c +++ b/drivers/scsi/qedi/qedi_iscsi.c @@ -1429,20 +1429,16 @@ static void qedi_cleanup_task(struct iscsi_task *task) cmd->scsi_cmd = NULL; } -static struct iscsi_transport_expand qedi_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - struct iscsi_transport qedi_iscsi_transport = { .owner = THIS_MODULE, .name = QEDI_MODULE_NAME, .caps = CAP_RECOVERY_L0 | CAP_HDRDGST | CAP_MULTI_R2T | CAP_DATADGST | - CAP_DATA_PATH_OFFLOAD | CAP_TEXT_NEGO | CAP_OPS_EXPAND, + CAP_DATA_PATH_OFFLOAD | CAP_TEXT_NEGO, .create_session = qedi_session_create, .destroy_session = qedi_session_destroy, .create_conn = qedi_conn_create, .bind_conn = qedi_conn_bind, - .ops_expand = &qedi_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .start_conn = qedi_conn_start, .stop_conn = iscsi_conn_stop, .destroy_conn = qedi_conn_destroy, diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 377d83762099..8d82d2a83059 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -246,24 +246,20 @@ static struct scsi_host_template qla4xxx_driver_template = { .vendor_id = SCSI_NL_VID_TYPE_PCI | PCI_VENDOR_ID_QLOGIC, }; -static struct iscsi_transport_expand qla4xxx_iscsi_expand = { - .unbind_conn = iscsi_conn_unbind, -}; - static struct iscsi_transport qla4xxx_iscsi_transport = { .owner = THIS_MODULE, .name = DRIVER_NAME, .caps = CAP_TEXT_NEGO | CAP_DATA_PATH_OFFLOAD | CAP_HDRDGST | CAP_DATADGST | CAP_LOGIN_OFFLOAD | - CAP_MULTI_R2T | CAP_OPS_EXPAND, + CAP_MULTI_R2T, .attr_is_visible = qla4_attr_is_visible, .create_session = qla4xxx_session_create, .destroy_session = qla4xxx_session_destroy, .start_conn = qla4xxx_conn_start, .create_conn = qla4xxx_conn_create, .bind_conn = qla4xxx_conn_bind, - .ops_expand = &qla4xxx_iscsi_expand, + .unbind_conn = iscsi_conn_unbind, .stop_conn = iscsi_conn_stop, .destroy_conn = qla4xxx_conn_destroy, .set_param = iscsi_set_param, diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index a213362524c9..517a3c4d879b 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -2257,11 +2257,7 @@ static void iscsi_ep_disconnect(struct iscsi_cls_conn *conn, bool is_active) ep = conn->ep; conn->ep = NULL; - if (session->transport->caps & CAP_OPS_EXPAND && - session->transport->ops_expand && - session->transport->ops_expand->unbind_conn) - session->transport->ops_expand->unbind_conn(conn, is_active); - + session->transport->unbind_conn(conn, is_active); session->transport->ep_disconnect(ep); ISCSI_DBG_TRANS_CONN(conn, "disconnect ep done.\n"); } @@ -3228,19 +3224,10 @@ iscsi_tgt_dscvr(struct iscsi_transport *transport, struct Scsi_Host *shost; struct sockaddr *dst_addr; int err; - int (*tgt_dscvr)(struct Scsi_Host *shost, enum iscsi_tgt_dscvr type, - uint32_t enable, struct sockaddr *dst_addr); - if (transport->caps & CAP_OPS_EXPAND) { - if (!transport->ops_expand || !transport->ops_expand->tgt_dscvr) - return -EINVAL; - tgt_dscvr = transport->ops_expand->tgt_dscvr; - } else { - if (!transport->ops_expand) - return -EINVAL; - tgt_dscvr = (int (*)(struct Scsi_Host *, enum iscsi_tgt_dscvr, uint32_t, - struct sockaddr *))(transport->ops_expand); - } + if (!transport->tgt_dscvr) + return -EINVAL; + shost = scsi_host_lookup(ev->u.tgt_dscvr.host_no); if (!shost) { printk(KERN_ERR "target discovery could not find host no %u\n", @@ -3250,8 +3237,8 @@ iscsi_tgt_dscvr(struct iscsi_transport *transport, dst_addr = (struct sockaddr *)((char*)ev + sizeof(*ev)); - err = tgt_dscvr(shost, ev->u.tgt_dscvr.type, - ev->u.tgt_dscvr.enable, dst_addr); + err = transport->tgt_dscvr(shost, ev->u.tgt_dscvr.type, + ev->u.tgt_dscvr.enable, dst_addr); scsi_host_put(shost); return err; } @@ -4904,10 +4891,8 @@ iscsi_register_transport(struct iscsi_transport *tt) int err; BUG_ON(!tt); - if (tt->caps & CAP_OPS_EXPAND) { - BUG_ON(!tt->ops_expand); - WARN_ON(tt->ep_disconnect && !tt->ops_expand->unbind_conn); - } + WARN_ON(tt->ep_disconnect && !tt->unbind_conn); + priv = iscsi_if_transport_lookup(tt); if (priv) return NULL; diff --git a/include/scsi/iscsi_if.h b/include/scsi/iscsi_if.h index 5679b9fb2b1e..5225a23f2d0e 100644 --- a/include/scsi/iscsi_if.h +++ b/include/scsi/iscsi_if.h @@ -761,7 +761,6 @@ enum iscsi_ping_status_code { and verification */ #define CAP_LOGIN_OFFLOAD 0x4000 /* offload session login */ -#define CAP_OPS_EXPAND 0x8000 /* oiscsi_transport->ops_expand flag */ /* * These flags describes reason of stop_conn() call */ diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h index 1f7574d89822..b7fd8b738aa8 100644 --- a/include/scsi/scsi_transport_iscsi.h +++ b/include/scsi/scsi_transport_iscsi.h @@ -29,15 +29,6 @@ struct bsg_job; struct iscsi_bus_flash_session; struct iscsi_bus_flash_conn; -/* - * The expansion of iscsi_transport to fix kabi while adding members. - */ -struct iscsi_transport_expand { - int (*tgt_dscvr)(struct Scsi_Host *shost, enum iscsi_tgt_dscvr type, - uint32_t enable, struct sockaddr *dst_addr); - void (*unbind_conn)(struct iscsi_cls_conn *conn, bool is_active); -}; - /** * struct iscsi_transport - iSCSI Transport template * @@ -91,6 +82,7 @@ struct iscsi_transport { void (*destroy_session) (struct iscsi_cls_session *session); struct iscsi_cls_conn *(*create_conn) (struct iscsi_cls_session *sess, uint32_t cid); + void (*unbind_conn) (struct iscsi_cls_conn *conn, bool is_active); int (*bind_conn) (struct iscsi_cls_session *session, struct iscsi_cls_conn *cls_conn, uint64_t transport_eph, int is_leading); @@ -132,15 +124,8 @@ struct iscsi_transport { int non_blocking); int (*ep_poll) (struct iscsi_endpoint *ep, int timeout_ms); void (*ep_disconnect) (struct iscsi_endpoint *ep); -#ifdef __GENKSYMS__ int (*tgt_dscvr) (struct Scsi_Host *shost, enum iscsi_tgt_dscvr type, uint32_t enable, struct sockaddr *dst_addr); -#else - /* - * onece ops_expand is used, caps must be set to CAP_OPS_EXPAND - */ - struct iscsi_transport_expand *ops_expand; -#endif int (*set_path) (struct Scsi_Host *shost, struct iscsi_path *params); int (*set_iface_param) (struct Scsi_Host *shost, void *data, uint32_t len); -- 2.20.1

1 1

[PATCH openEuler-1.0-LTS] drm/amdkfd: Check for null pointer after calling kmemdup
by Yongqiang Liu 22 Dec '22

22 Dec '22

From: Jiasheng Jiang <jiasheng(a)iscas.ac.cn> stable inclusion from stable-v6.0.14 commit abfaf0eee97925905e742aa3b0b72e04a918fa9e category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I6694U CVE: CVE-2022-3108 -------------------------------- As the possible failure of the allocation, kmemdup() may return NULL pointer. Therefore, it should be better to check the 'props2' in order to prevent the dereference of NULL pointer. Fixes: 3a87177eb141 ("drm/amdkfd: Add topology support for dGPUs") Signed-off-by: Jiasheng Jiang <jiasheng(a)iscas.ac.cn> Reviewed-by: Felix Kuehling <Felix.Kuehling(a)amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Ren Zhijie <renzhijie2(a)huawei.com> Reviewed-by: songping yu <yusongping(a)huawei.com> Reviewed-by: Chen Hui <judy.chenhui(a)huawei.com> Reviewed-by: Xiu Jianfeng <xiujianfeng(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- drivers/gpu/drm/amd/amdkfd/kfd_crat.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index ee4996029a86..e84285130bde 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -397,6 +397,9 @@ static int kfd_parse_subtype_iolink(struct crat_subtype_iolink *iolink, return -ENODEV; /* same everything but the other direction */ props2 = kmemdup(props, sizeof(*props2), GFP_KERNEL); + if (!props2) + return -ENOMEM; + props2->node_from = id_to; props2->node_to = id_from; props2->kobj = NULL; -- 2.25.1

1 0

[PATCH openEuler-1.0-LTS 1/4] Bluetooth: L2CAP: Fix use-after-free caused by l2cap_chan_put
by Yongqiang Liu 19 Dec '22

19 Dec '22

From: Luiz Augusto von Dentz <luiz.von.dentz(a)intel.com> stable inclusion from stable-v4.19.255 commit d0be8347c623e0ac4202a1d4e0373882821f56b0 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I64WCC CVE: CVE-2022-20566 -------------------------------- commit d0be8347c623e0ac4202a1d4e0373882821f56b0 upstream. This fixes the following trace which is caused by hci_rx_work starting up *after* the final channel reference has been put() during sock_close() but *before* the references to the channel have been destroyed, so instead the code now rely on kref_get_unless_zero/l2cap_chan_hold_unless_zero to prevent referencing a channel that is about to be destroyed. refcount_t: increment on 0; use-after-free. BUG: KASAN: use-after-free in refcount_dec_and_test+0x20/0xd0 Read of size 4 at addr ffffffc114f5bf18 by task kworker/u17:14/705 CPU: 4 PID: 705 Comm: kworker/u17:14 Tainted: G S W 4.14.234-00003-g1fb6d0bd49a4-dirty #28 Hardware name: Qualcomm Technologies, Inc. SM8150 V2 PM8150 Google Inc. MSM sm8150 Flame DVT (DT) Workqueue: hci0 hci_rx_work Call trace: dump_backtrace+0x0/0x378 show_stack+0x20/0x2c dump_stack+0x124/0x148 print_address_description+0x80/0x2e8 __kasan_report+0x168/0x188 kasan_report+0x10/0x18 __asan_load4+0x84/0x8c refcount_dec_and_test+0x20/0xd0 l2cap_chan_put+0x48/0x12c l2cap_recv_frame+0x4770/0x6550 l2cap_recv_acldata+0x44c/0x7a4 hci_acldata_packet+0x100/0x188 hci_rx_work+0x178/0x23c process_one_work+0x35c/0x95c worker_thread+0x4cc/0x960 kthread+0x1a8/0x1c4 ret_from_fork+0x10/0x18 Cc: stable(a)kernel.org Reported-by: Lee Jones <lee.jones(a)linaro.org> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz(a)intel.com> Tested-by: Lee Jones <lee.jones(a)linaro.org> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz(a)intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Lu Jialin <lujialin4(a)huawei.com> Reviewed-by: Xiu Jianfeng <xiujianfeng(a)huawei.com> Reviewed-by: Wang Weiyang <wangweiyang2(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- include/net/bluetooth/l2cap.h | 1 + net/bluetooth/l2cap_core.c | 61 +++++++++++++++++++++++++++-------- 2 files changed, 49 insertions(+), 13 deletions(-) diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h index 21dbd38f724d..da0ef935c5a9 100644 --- a/include/net/bluetooth/l2cap.h +++ b/include/net/bluetooth/l2cap.h @@ -798,6 +798,7 @@ enum { }; void l2cap_chan_hold(struct l2cap_chan *c); +struct l2cap_chan *l2cap_chan_hold_unless_zero(struct l2cap_chan *c); void l2cap_chan_put(struct l2cap_chan *c); static inline void l2cap_chan_lock(struct l2cap_chan *chan) diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 3b02e881cec1..7c2bb0571573 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -113,7 +113,8 @@ static struct l2cap_chan *__l2cap_get_chan_by_scid(struct l2cap_conn *conn, } /* Find channel with given SCID. - * Returns locked channel. */ + * Returns a reference locked channel. + */ static struct l2cap_chan *l2cap_get_chan_by_scid(struct l2cap_conn *conn, u16 cid) { @@ -121,15 +122,19 @@ static struct l2cap_chan *l2cap_get_chan_by_scid(struct l2cap_conn *conn, mutex_lock(&conn->chan_lock); c = __l2cap_get_chan_by_scid(conn, cid); - if (c) - l2cap_chan_lock(c); + if (c) { + /* Only lock if chan reference is not 0 */ + c = l2cap_chan_hold_unless_zero(c); + if (c) + l2cap_chan_lock(c); + } mutex_unlock(&conn->chan_lock); return c; } /* Find channel with given DCID. - * Returns locked channel. + * Returns a reference locked channel. */ static struct l2cap_chan *l2cap_get_chan_by_dcid(struct l2cap_conn *conn, u16 cid) @@ -138,8 +143,12 @@ static struct l2cap_chan *l2cap_get_chan_by_dcid(struct l2cap_conn *conn, mutex_lock(&conn->chan_lock); c = __l2cap_get_chan_by_dcid(conn, cid); - if (c) - l2cap_chan_lock(c); + if (c) { + /* Only lock if chan reference is not 0 */ + c = l2cap_chan_hold_unless_zero(c); + if (c) + l2cap_chan_lock(c); + } mutex_unlock(&conn->chan_lock); return c; @@ -164,8 +173,12 @@ static struct l2cap_chan *l2cap_get_chan_by_ident(struct l2cap_conn *conn, mutex_lock(&conn->chan_lock); c = __l2cap_get_chan_by_ident(conn, ident); - if (c) - l2cap_chan_lock(c); + if (c) { + /* Only lock if chan reference is not 0 */ + c = l2cap_chan_hold_unless_zero(c); + if (c) + l2cap_chan_lock(c); + } mutex_unlock(&conn->chan_lock); return c; @@ -486,6 +499,16 @@ void l2cap_chan_hold(struct l2cap_chan *c) kref_get(&c->kref); } +struct l2cap_chan *l2cap_chan_hold_unless_zero(struct l2cap_chan *c) +{ + BT_DBG("chan %p orig refcnt %u", c, kref_read(&c->kref)); + + if (!kref_get_unless_zero(&c->kref)) + return NULL; + + return c; +} + void l2cap_chan_put(struct l2cap_chan *c) { BT_DBG("chan %p orig refcnt %d", c, kref_read(&c->kref)); @@ -1796,7 +1819,10 @@ static struct l2cap_chan *l2cap_global_chan_by_psm(int state, __le16 psm, src_match = !bacmp(&c->src, src); dst_match = !bacmp(&c->dst, dst); if (src_match && dst_match) { - l2cap_chan_hold(c); + c = l2cap_chan_hold_unless_zero(c); + if (!c) + continue; + read_unlock(&chan_list_lock); return c; } @@ -1811,7 +1837,7 @@ static struct l2cap_chan *l2cap_global_chan_by_psm(int state, __le16 psm, } if (c1) - l2cap_chan_hold(c1); + c1 = l2cap_chan_hold_unless_zero(c1); read_unlock(&chan_list_lock); @@ -4198,6 +4224,7 @@ static inline int l2cap_config_req(struct l2cap_conn *conn, unlock: l2cap_chan_unlock(chan); + l2cap_chan_put(chan); return err; } @@ -4310,6 +4337,7 @@ static inline int l2cap_config_rsp(struct l2cap_conn *conn, done: l2cap_chan_unlock(chan); + l2cap_chan_put(chan); return err; } @@ -5037,6 +5065,7 @@ static inline int l2cap_move_channel_req(struct l2cap_conn *conn, l2cap_send_move_chan_rsp(chan, result); l2cap_chan_unlock(chan); + l2cap_chan_put(chan); return 0; } @@ -5129,6 +5158,7 @@ static void l2cap_move_continue(struct l2cap_conn *conn, u16 icid, u16 result) } l2cap_chan_unlock(chan); + l2cap_chan_put(chan); } static void l2cap_move_fail(struct l2cap_conn *conn, u8 ident, u16 icid, @@ -5158,6 +5188,7 @@ static void l2cap_move_fail(struct l2cap_conn *conn, u8 ident, u16 icid, l2cap_send_move_chan_cfm(chan, L2CAP_MC_UNCONFIRMED); l2cap_chan_unlock(chan); + l2cap_chan_put(chan); } static int l2cap_move_channel_rsp(struct l2cap_conn *conn, @@ -5221,6 +5252,7 @@ static int l2cap_move_channel_confirm(struct l2cap_conn *conn, l2cap_send_move_chan_cfm_rsp(conn, cmd->ident, icid); l2cap_chan_unlock(chan); + l2cap_chan_put(chan); return 0; } @@ -5256,6 +5288,7 @@ static inline int l2cap_move_channel_confirm_rsp(struct l2cap_conn *conn, } l2cap_chan_unlock(chan); + l2cap_chan_put(chan); return 0; } @@ -5641,12 +5674,11 @@ static inline int l2cap_le_credits(struct l2cap_conn *conn, if (credits > max_credits) { BT_ERR("LE credits overflow"); l2cap_send_disconn_req(chan, ECONNRESET); - l2cap_chan_unlock(chan); /* Return 0 so that we don't trigger an unnecessary * command reject packet. */ - return 0; + goto unlock; } chan->tx_credits += credits; @@ -5657,7 +5689,9 @@ static inline int l2cap_le_credits(struct l2cap_conn *conn, if (chan->tx_credits) chan->ops->resume(chan); +unlock: l2cap_chan_unlock(chan); + l2cap_chan_put(chan); return 0; } @@ -6989,6 +7023,7 @@ static void l2cap_data_channel(struct l2cap_conn *conn, u16 cid, done: l2cap_chan_unlock(chan); + l2cap_chan_put(chan); } static void l2cap_conless_channel(struct l2cap_conn *conn, __le16 psm, @@ -7393,7 +7428,7 @@ static struct l2cap_chan *l2cap_global_fixed_chan(struct l2cap_chan *c, if (src_type != c->src_type) continue; - l2cap_chan_hold(c); + c = l2cap_chan_hold_unless_zero(c); read_unlock(&chan_list_lock); return c; } -- 2.25.1

1 3

[PATCH openEuler-5.10] cpuidle: add cpuidle-haltpoll driver module parameter
by Zheng Zengkai 19 Dec '22

19 Dec '22

From: Xiangyou Xie <xiexiangyou(a)huawei.com> hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I6404X?from=project-issue CVE: NA --------------------------- To ensure energy efficiency, haltpoll is disabled by default. But In some performance scenarios, you can enable haltpoll using the following methods: echo Y > /sys/module/cpuidle_haltpoll/parameters/force Signed-off-by: Xiangyou Xie <xiexiangyou(a)huawei.com> Reviewed-by: Keqian Zhu <zhukeqian1(a)huawei.com> Reviewed-by: Qingqing Li <liqingqing3(a)huawei.com> Reviewed-by: chenhui <judy.chenhui(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> --- drivers/cpuidle/cpuidle-haltpoll.c | 97 +++++++++++++++++++++++----- drivers/cpuidle/governors/haltpoll.c | 4 +- 2 files changed, 82 insertions(+), 19 deletions(-) diff --git a/drivers/cpuidle/cpuidle-haltpoll.c b/drivers/cpuidle/cpuidle-haltpoll.c index e22f7867609b..902af11bd0e0 100644 --- a/drivers/cpuidle/cpuidle-haltpoll.c +++ b/drivers/cpuidle/cpuidle-haltpoll.c @@ -18,9 +18,17 @@ #include <linux/kvm_para.h> #include <linux/cpuidle_haltpoll.h> -static bool force __read_mostly; -module_param(force, bool, 0444); -MODULE_PARM_DESC(force, "Load unconditionally"); +static bool force; +MODULE_PARM_DESC(force, "bool, enable haltpoll driver"); +static int enable_haltpoll_driver(const char *val, const struct kernel_param *kp); +static int register_haltpoll_driver(void); +static void unregister_haltpoll_driver(void); + +static const struct kernel_param_ops enable_haltpoll_ops = { + .set = enable_haltpoll_driver, + .get = param_get_bool, +}; +module_param_cb(force, &enable_haltpoll_ops, &force, 0644); static struct cpuidle_device __percpu *haltpoll_cpuidle_devices; static enum cpuhp_state haltpoll_hp_state; @@ -36,6 +44,42 @@ static int default_enter_idle(struct cpuidle_device *dev, return index; } + +static int enable_haltpoll_driver(const char *val, const struct kernel_param *kp) +{ +#ifdef CONFIG_ARM64 + int ret; + bool do_enable; + + if (!val) + return 0; + + ret = strtobool(val, &do_enable); + + if (ret || force == do_enable) + return ret; + + if (do_enable) { + ret = register_haltpoll_driver(); + + if (!ret) { + pr_info("Enable haltpoll driver.\n"); + force = 1; + } else { + pr_err("Fail to enable haltpoll driver.\n"); + } + } else { + unregister_haltpoll_driver(); + force = 0; + pr_info("Unregister haltpoll driver.\n"); + } + + return ret; +#else + return -1; +#endif +} + static struct cpuidle_driver haltpoll_driver = { .name = "haltpoll", .governor = "haltpoll", @@ -84,22 +128,18 @@ static int haltpoll_cpu_offline(unsigned int cpu) return 0; } -static void haltpoll_uninit(void) -{ - if (haltpoll_hp_state) - cpuhp_remove_state(haltpoll_hp_state); - cpuidle_unregister_driver(&haltpoll_driver); - - free_percpu(haltpoll_cpuidle_devices); - haltpoll_cpuidle_devices = NULL; -} static bool haltpoll_want(void) { return kvm_para_has_hint(KVM_HINTS_REALTIME); } -static int __init haltpoll_init(void) +static void haltpoll_uninit(void) +{ + unregister_haltpoll_driver(); +} + +static int register_haltpoll_driver(void) { int ret; struct cpuidle_driver *drv = &haltpoll_driver; @@ -112,9 +152,6 @@ static int __init haltpoll_init(void) cpuidle_poll_state_init(drv); - if (!force && (!kvm_para_available() || !haltpoll_want())) - return -ENODEV; - ret = cpuidle_register_driver(drv); if (ret < 0) return ret; @@ -137,9 +174,35 @@ static int __init haltpoll_init(void) return ret; } +static void unregister_haltpoll_driver(void) +{ + if (haltpoll_hp_state) + cpuhp_remove_state(haltpoll_hp_state); + cpuidle_unregister_driver(&haltpoll_driver); + + free_percpu(haltpoll_cpuidle_devices); + haltpoll_cpuidle_devices = NULL; + +} + +static int __init haltpoll_init(void) +{ + int ret = 0; +#ifdef CONFIG_X86 + /* Do not load haltpoll if idle= is passed */ + if (boot_option_idle_override != IDLE_NO_OVERRIDE) + return -ENODEV; +#endif + if (force || (haltpoll_want() && kvm_para_available())) + ret = register_haltpoll_driver(); + + return ret; +} + static void __exit haltpoll_exit(void) { - haltpoll_uninit(); + if (haltpoll_cpuidle_devices) + haltpoll_uninit(); } module_init(haltpoll_init); diff --git a/drivers/cpuidle/governors/haltpoll.c b/drivers/cpuidle/governors/haltpoll.c index 4e814683509e..415004793bf1 100644 --- a/drivers/cpuidle/governors/haltpoll.c +++ b/drivers/cpuidle/governors/haltpoll.c @@ -39,7 +39,7 @@ module_param(guest_halt_poll_grow_start, uint, 0644); static bool guest_halt_poll_allow_shrink __read_mostly = true; module_param(guest_halt_poll_allow_shrink, bool, 0644); -static bool enable __read_mostly; +static bool enable __read_mostly = true; module_param(enable, bool, 0444); MODULE_PARM_DESC(enable, "Load unconditionally"); @@ -144,7 +144,7 @@ static struct cpuidle_governor haltpoll_governor = { static int __init init_haltpoll(void) { - if (kvm_para_available() || enable) + if (enable) return cpuidle_register_governor(&haltpoll_governor); return 0; -- 2.20.1

1 0

[PATCH openEuler-5.10 1/2] block: fix crash on cmpxchg for request_wrapper
by Zheng Zengkai 19 Dec '22

19 Dec '22

From: Yu Kuai <yukuai3(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I65K8D CVE: NA -------------------------------- Now that address of request_wrapper is caculated by address of request plus cmd_size, if cmd_size is not aligned to 8 bytes, request_wrapper will end up not aligned to 8 bytes as well, which will crash in arm64 because assembly instruction casal requires that operand address is aligned to 8 bytes: Internal error: Oops: 96000021 [#1] SMP pc : blk_account_io_latency+0x54/0x134 Call trace: blk_account_io_latency+0x54/0x134 blk_account_io_done+0x3c/0x4c __blk_mq_end_request+0x78/0x134 scsi_end_request+0xcc/0x1f0 scsi_io_completion+0x88/0x240 scsi_finish_command+0x104/0x140 scsi_softirq_done+0x90/0x180 blk_mq_complete_request+0x5c/0x70 scsi_mq_done+0x4c/0x100 Fix the problem by declaring request_wrapper as aligned to cachline, and placing it before request. Fixes: 82327165da5c ("blk-mq: don't access request_wrapper if request is not allocated from block layer") Signed-off-by: Yu Kuai <yukuai3(a)huawei.com> Reviewed-by: Hou Tao <houtao1(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> --- block/blk-flush.c | 8 +++++--- block/blk-mq.c | 2 +- block/blk-mq.h | 9 +++------ 3 files changed, 9 insertions(+), 10 deletions(-) diff --git a/block/blk-flush.c b/block/blk-flush.c index 65753f781c20..093c581a2651 100644 --- a/block/blk-flush.c +++ b/block/blk-flush.c @@ -470,6 +470,7 @@ struct blk_flush_queue *blk_alloc_flush_queue(int node, int cmd_size, gfp_t flags) { struct blk_flush_queue *fq; + struct request_wrapper *wrapper; int rq_sz = sizeof(struct request) + sizeof(struct request_wrapper); fq = kzalloc_node(sizeof(*fq), flags, node); @@ -479,10 +480,11 @@ struct blk_flush_queue *blk_alloc_flush_queue(int node, int cmd_size, spin_lock_init(&fq->mq_flush_lock); rq_sz = round_up(rq_sz + cmd_size, cache_line_size()); - fq->flush_rq = kzalloc_node(rq_sz, flags, node); - if (!fq->flush_rq) + wrapper = kzalloc_node(rq_sz, flags, node); + if (!wrapper) goto fail_rq; + fq->flush_rq = (struct request *)(wrapper + 1); INIT_LIST_HEAD(&fq->flush_queue[0]); INIT_LIST_HEAD(&fq->flush_queue[1]); INIT_LIST_HEAD(&fq->flush_data_in_flight); @@ -501,7 +503,7 @@ void blk_free_flush_queue(struct blk_flush_queue *fq) if (!fq) return; - kfree(fq->flush_rq); + kfree(request_to_wrapper(fq->flush_rq)); kfree(fq); } diff --git a/block/blk-mq.c b/block/blk-mq.c index 1c4a4e197e65..ffabe9c3de31 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -2642,7 +2642,7 @@ static int blk_mq_alloc_rqs(struct blk_mq_tag_set *set, to_do = min(entries_per_page, depth - i); left -= to_do * rq_size; for (j = 0; j < to_do; j++) { - struct request *rq = p; + struct request *rq = p + sizeof(struct request_wrapper); tags->static_rqs[i] = rq; if (blk_mq_init_request(set, rq, hctx_idx, node)) { diff --git a/block/blk-mq.h b/block/blk-mq.h index 6254abe9c112..dcb2077e4db6 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -40,14 +40,11 @@ struct blk_mq_ctx { struct request_wrapper { /* Time that I/O was counted in part_get_stat_info(). */ u64 stat_time_ns; -}; +} ____cacheline_aligned_in_smp; -static inline struct request_wrapper *request_to_wrapper(struct request *rq) +static inline struct request_wrapper *request_to_wrapper(void *rq) { - unsigned long addr = (unsigned long)rq; - - addr += sizeof(*rq) + rq->q->tag_set->cmd_size; - return (struct request_wrapper *)addr; + return rq - sizeof(struct request_wrapper); } void blk_mq_exit_queue(struct request_queue *q); -- 2.20.1

1 1

[PATCH openEuler-5.10 1/7] Bluetooth: L2CAP: Fix build errors in some archs
by Zheng Zengkai 19 Dec '22

19 Dec '22

From: Luiz Augusto von Dentz <luiz.von.dentz(a)intel.com> stable inclusion from stable-v5.10.141 commit 38267d266336a7fb9eae9be23567a44776c6e4ca category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I64WCC CVE: CVE-2022-20566 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… ------------------------------- commit b840304fb46cdf7012722f456bce06f151b3e81b upstream. This attempts to fix the follow errors: In function 'memcmp', inlined from 'bacmp' at ./include/net/bluetooth/bluetooth.h:347:9, inlined from 'l2cap_global_chan_by_psm' at net/bluetooth/l2cap_core.c:2003:15: ./include/linux/fortify-string.h:44:33: error: '__builtin_memcmp' specified bound 6 exceeds source size 0 [-Werror=stringop-overread] 44 | #define __underlying_memcmp __builtin_memcmp | ^ ./include/linux/fortify-string.h:420:16: note: in expansion of macro '__underlying_memcmp' 420 | return __underlying_memcmp(p, q, size); | ^~~~~~~~~~~~~~~~~~~ In function 'memcmp', inlined from 'bacmp' at ./include/net/bluetooth/bluetooth.h:347:9, inlined from 'l2cap_global_chan_by_psm' at net/bluetooth/l2cap_core.c:2004:15: ./include/linux/fortify-string.h:44:33: error: '__builtin_memcmp' specified bound 6 exceeds source size 0 [-Werror=stringop-overread] 44 | #define __underlying_memcmp __builtin_memcmp | ^ ./include/linux/fortify-string.h:420:16: note: in expansion of macro '__underlying_memcmp' 420 | return __underlying_memcmp(p, q, size); | ^~~~~~~~~~~~~~~~~~~ Fixes: 332f1795ca20 ("Bluetooth: L2CAP: Fix l2cap_global_chan_by_psm regression") Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz(a)intel.com> Cc: Sudip Mukherjee <sudipm.mukherjee(a)gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Lu Jialin <lujialin4(a)huawei.com> Reviewed-by: Xiu Jianfeng <xiujianfeng(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> --- net/bluetooth/l2cap_core.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 70bfd9e8913e..f78ad8f536f7 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -1988,11 +1988,11 @@ static struct l2cap_chan *l2cap_global_chan_by_psm(int state, __le16 psm, src_match = !bacmp(&c->src, src); dst_match = !bacmp(&c->dst, dst); if (src_match && dst_match) { - c = l2cap_chan_hold_unless_zero(c); - if (c) { - read_unlock(&chan_list_lock); - return c; - } + if (!l2cap_chan_hold_unless_zero(c)) + continue; + + read_unlock(&chan_list_lock); + return c; } /* Closest match */ -- 2.20.1

1 6

[PATCH openEuler-1.0-LTS] hv_netvsc: Add check for kvmalloc_array
by Yongqiang Liu 19 Dec '22

19 Dec '22

From: Jiasheng Jiang <jiasheng(a)iscas.ac.cn> mainline inclusion from mainline-v5.17 commit 886e44c9298a6b428ae046e2fa092ca52e822e6a category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I6694H CVE: CVE-2022-3107 -------------------------------- As the potential failure of the kvmalloc_array(), it should be better to check and restore the 'data' if fails in order to avoid the dereference of the NULL pointer. Fixes: 6ae746711263 ("hv_netvsc: Add per-cpu ethtool stats for netvsc") Signed-off-by: Jiasheng Jiang <jiasheng(a)iscas.ac.cn> Link: https://lore.kernel.org/r/20220314020125.2365084-1-jiasheng@iscas.ac.cn Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Baisong Zhong <zhongbaisong(a)huawei.com> Reviewed-by: Liu Jian <liujian56(a)huawei.com> Reviewed-by: Yue Haibing <yuehaibing(a)huawei.com> Reviewed-by: Xiu Jianfeng <xiujianfeng(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- drivers/net/hyperv/netvsc_drv.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c index 1f9f7fcdb0eb..0de42e8ec2bb 100644 --- a/drivers/net/hyperv/netvsc_drv.c +++ b/drivers/net/hyperv/netvsc_drv.c @@ -1453,6 +1453,9 @@ static void netvsc_get_ethtool_stats(struct net_device *dev, pcpu_sum = kvmalloc_array(num_possible_cpus(), sizeof(struct netvsc_ethtool_pcpu_stats), GFP_KERNEL); + if (!pcpu_sum) + return; + netvsc_get_pcpu_stats(dev, pcpu_sum); for_each_present_cpu(cpu) { struct netvsc_ethtool_pcpu_stats *this_sum = &pcpu_sum[cpu]; -- 2.25.1

1 0

[PATCH openEuler-5.10 01/10] sched/qos: Don't unthrottle cfs_rq when cfs_rq is throttled by qos
by Zheng Zengkai 16 Dec '22

16 Dec '22

From: Zhang Qiao <zhangqiao22(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I64OUS CVE: NA ------------------------------- When a cfs_rq throttled by qos, mark cfs_rq->throttled as 1, and cfs bw will unthrottled this cfs_rq by mistake, it cause a list_del_valid warning. So add macro QOS_THROTTLED(=2), when a cfs_rq is throttled by qos, we mark the cfs_rq->throttled as QOS_THROTTLED, will check the value of cfs_rq->throttled before unthrottle a cfs_rq. Signed-off-by: Zhang Qiao <zhangqiao22(a)huawei.com> Reviewed-by: Chen Hui <judy.chenhui(a)huawei.com> Reviewed-by: zheng zucheng <zhengzucheng(a)huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> --- kernel/sched/fair.c | 102 +++++++++++++++++++++++++++++--------------- 1 file changed, 68 insertions(+), 34 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 654e964b5c31..d3c4b945c019 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -124,6 +124,13 @@ int __weak arch_asym_cpu_priority(int cpu) #endif #ifdef CONFIG_QOS_SCHED + +/* + * To distinguish cfs bw, use QOS_THROTTLED mark cfs_rq->throttled + * when qos throttled(and cfs bw throttle mark cfs_rq->throttled as 1). + */ +#define QOS_THROTTLED 2 + static DEFINE_PER_CPU_SHARED_ALIGNED(struct list_head, qos_throttled_cfs_rq); static DEFINE_PER_CPU_SHARED_ALIGNED(struct hrtimer, qos_overload_timer); static DEFINE_PER_CPU(int, qos_cpu_overload); @@ -4932,6 +4939,14 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) se = cfs_rq->tg->se[cpu_of(rq)]; +#ifdef CONFIG_QOS_SCHED + /* + * if this cfs_rq throttled by qos, not need unthrottle it. + */ + if (cfs_rq->throttled == QOS_THROTTLED) + return; +#endif + cfs_rq->throttled = 0; update_rq_clock(rq); @@ -7278,26 +7293,6 @@ static inline bool is_offline_task(struct task_struct *p) static void start_qos_hrtimer(int cpu); -static int qos_tg_unthrottle_up(struct task_group *tg, void *data) -{ - struct rq *rq = data; - struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; - - cfs_rq->throttle_count--; - - return 0; -} - -static int qos_tg_throttle_down(struct task_group *tg, void *data) -{ - struct rq *rq = data; - struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; - - cfs_rq->throttle_count++; - - return 0; -} - static void throttle_qos_cfs_rq(struct cfs_rq *cfs_rq) { struct rq *rq = rq_of(cfs_rq); @@ -7309,7 +7304,7 @@ static void throttle_qos_cfs_rq(struct cfs_rq *cfs_rq) /* freeze hierarchy runnable averages while throttled */ rcu_read_lock(); - walk_tg_tree_from(cfs_rq->tg, qos_tg_throttle_down, tg_nop, (void *)rq); + walk_tg_tree_from(cfs_rq->tg, tg_throttle_down, tg_nop, (void *)rq); rcu_read_unlock(); task_delta = cfs_rq->h_nr_running; @@ -7320,8 +7315,13 @@ static void throttle_qos_cfs_rq(struct cfs_rq *cfs_rq) if (!se->on_rq) break; - if (dequeue) + if (dequeue) { dequeue_entity(qcfs_rq, se, DEQUEUE_SLEEP); + } else { + update_load_avg(qcfs_rq, se, 0); + se_update_runnable(se); + } + qcfs_rq->h_nr_running -= task_delta; qcfs_rq->idle_h_nr_running -= idle_task_delta; @@ -7339,7 +7339,7 @@ static void throttle_qos_cfs_rq(struct cfs_rq *cfs_rq) if (list_empty(&per_cpu(qos_throttled_cfs_rq, cpu_of(rq)))) start_qos_hrtimer(cpu_of(rq)); - cfs_rq->throttled = 1; + cfs_rq->throttled = QOS_THROTTLED; list_add(&cfs_rq->qos_throttled_list, &per_cpu(qos_throttled_cfs_rq, cpu_of(rq))); @@ -7349,12 +7349,14 @@ static void unthrottle_qos_cfs_rq(struct cfs_rq *cfs_rq) { struct rq *rq = rq_of(cfs_rq); struct sched_entity *se; - int enqueue = 1; unsigned int prev_nr = cfs_rq->h_nr_running; long task_delta, idle_task_delta; se = cfs_rq->tg->se[cpu_of(rq)]; + if (cfs_rq->throttled != QOS_THROTTLED) + return; + cfs_rq->throttled = 0; update_rq_clock(rq); @@ -7362,7 +7364,7 @@ static void unthrottle_qos_cfs_rq(struct cfs_rq *cfs_rq) /* update hierarchical throttle state */ rcu_read_lock(); - walk_tg_tree_from(cfs_rq->tg, tg_nop, qos_tg_unthrottle_up, (void *)rq); + walk_tg_tree_from(cfs_rq->tg, tg_nop, tg_unthrottle_up, (void *)rq); rcu_read_unlock(); if (!cfs_rq->load.weight) @@ -7372,26 +7374,58 @@ static void unthrottle_qos_cfs_rq(struct cfs_rq *cfs_rq) idle_task_delta = cfs_rq->idle_h_nr_running; for_each_sched_entity(se) { if (se->on_rq) - enqueue = 0; + break; cfs_rq = cfs_rq_of(se); - if (enqueue) - enqueue_entity(cfs_rq, se, ENQUEUE_WAKEUP); + enqueue_entity(cfs_rq, se, ENQUEUE_WAKEUP); + cfs_rq->h_nr_running += task_delta; cfs_rq->idle_h_nr_running += idle_task_delta; if (cfs_rq_throttled(cfs_rq)) - break; + goto unthrottle_throttle; } - assert_list_leaf_cfs_rq(rq); + for_each_sched_entity(se) { + cfs_rq = cfs_rq_of(se); - if (!se) { - add_nr_running(rq, task_delta); - if (prev_nr < 2 && prev_nr + task_delta >= 2) - overload_set(rq); + update_load_avg(cfs_rq, se, UPDATE_TG); + se_update_runnable(se); + + cfs_rq->h_nr_running += task_delta; + cfs_rq->idle_h_nr_running += idle_task_delta; + + /* end evaluation on encountering a throttled cfs_rq */ + if (cfs_rq_throttled(cfs_rq)) + goto unthrottle_throttle; + + /* + * One parent has been throttled and cfs_rq removed from the + * list. Add it back to not break the leaf list. + */ + if (throttled_hierarchy(cfs_rq)) + list_add_leaf_cfs_rq(cfs_rq); + } + + add_nr_running(rq, task_delta); + if (prev_nr < 2 && prev_nr + task_delta >= 2) + overload_set(rq); + +unthrottle_throttle: + /* + * The cfs_rq_throttled() breaks in the above iteration can result in + * incomplete leaf list maintenance, resulting in triggering the + * assertion below. + */ + for_each_sched_entity(se) { + cfs_rq = cfs_rq_of(se); + + if (list_add_leaf_cfs_rq(cfs_rq)) + break; } + assert_list_leaf_cfs_rq(rq); + /* Determine whether we need to wake up potentially idle CPU: */ if (rq->curr == rq->idle && rq->cfs.nr_running) resched_curr(rq); -- 2.20.1

1 9

[PATCH openEuler-5.10-LTS 1/7] efi/libstub: Fix build error in efi-stub.c for riscv
by Zheng Zengkai 16 Dec '22

16 Dec '22

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I66G0M ----------------------------------------------- Commit 36a1a8916de5 ("efi/loongarch: Add efistub booting support") introduced the following macro definition for riscv in file drivers/firmware/efi/libstub/efi-stub.c # define EFI_RT_VIRTUAL_LIMIT DEFAULT_MAP_WINDOW_64 # define EFI_RT_VIRTUAL_LIMIT TASK_SIZE_MIN # define EFI_RT_VIRTUAL_LIMIT TASK_SIZE but related patches have not been merged for riscv, like e8a62cc26ddf ("riscv: Implement sv48 support"), 01abdfeac81b ("riscv: compat: Support TASK_SIZE for compat mode"), Causing build error in riscv: build failed: riscv, allmodconfig <line too long ...> n this function); did you mean ‘TASK_SIZE_MAX’? # define EFI_RT_VIRTUAL_LIMIT TASK_SIZE_MIN ^ drivers/firmware/efi/libstub/efi-stub.c:289:31: note: in expansion of macro ‘EFI_RT_VIRTUAL_LIMIT’ static const u64 headroom = EFI_RT_VIRTUAL_LIMIT - ^~~~~~~~~~~~~~~~~~~~ drivers/firmware/efi/libstub/efi-stub.c:44:32: note: each undeclared identifier is reported only once for each function it appears in # define EFI_RT_VIRTUAL_LIMIT TASK_SIZE_MIN ^ drivers/firmware/efi/libstub/efi-stub.c:289:31: note: in expansion of macro ‘EFI_RT_VIRTUAL_LIMIT’ static const u64 headroom = EFI_RT_VIRTUAL_LIMIT - ^~~~~~~~~~~~~~~~~~~~ make[4]: *** [drivers/firmware/efi/libstub/efi-stub.o] Error 1 make[3]: *** [drivers/firmware/efi/libstub] Error 2 make[2]: *** [drivers/firmware/efi] Error 2 make[1]: *** [drivers/firmware] Error 2 make[1]: *** Waiting for unfinished jobs.... make: *** [drivers] Error 2 Fix it by removing the CONFIG_RISCV condition for EFI_RT_VIRTUAL_LIMIT definition. Fixes: 36a1a8916de5 ("efi/loongarch: Add efistub booting support") Signed-off-by: Zheng Zengkai <zhengzengkai(a)huawei.com> Reviewed-by: Wei Li <liwei391(a)huawei.com> --- drivers/firmware/efi/libstub/efi-stub.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/firmware/efi/libstub/efi-stub.c b/drivers/firmware/efi/libstub/efi-stub.c index 8e0f64b3db69..96129f0fc60e 100644 --- a/drivers/firmware/efi/libstub/efi-stub.c +++ b/drivers/firmware/efi/libstub/efi-stub.c @@ -40,7 +40,7 @@ #ifdef CONFIG_ARM64 # define EFI_RT_VIRTUAL_LIMIT DEFAULT_MAP_WINDOW_64 -#elif defined(CONFIG_RISCV) || defined(CONFIG_LOONGARCH) +#elif defined(CONFIG_LOONGARCH) # define EFI_RT_VIRTUAL_LIMIT TASK_SIZE_MIN #else # define EFI_RT_VIRTUAL_LIMIT TASK_SIZE -- 2.20.1

1 6

[PATCH openEuler-1.0-LTS 1/3] xen/netback: Ensure protocol headers don't fall in the non-linear area
by Yongqiang Liu 16 Dec '22

16 Dec '22

From: Ross Lagerwall <ross.lagerwall(a)citrix.com> mainline inclusion from mainline-v6.1 commit ad7f402ae4f466647c3a669b8a6f3e5d4271c84a category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I651EB CVE: CVE-2022-3643 -------------------------------- In some cases, the frontend may send a packet where the protocol headers are spread across multiple slots. This would result in netback creating an skb where the protocol headers spill over into the non-linear area. Some drivers and NICs don't handle this properly resulting in an interface reset or worse. This issue was introduced by the removal of an unconditional skb pull in the tx path to improve performance. Fix this without reintroducing the pull by setting up grant copy ops for as many slots as needed to reach the XEN_NETBACK_TX_COPY_LEN size. Adjust the rest of the code to handle multiple copy operations per skb. This is XSA-423 / CVE-2022-3643. Fixes: 7e5d7753956b ("xen-netback: remove unconditional __pskb_pull_tail() in guest Tx path") Signed-off-by: Ross Lagerwall <ross.lagerwall(a)citrix.com> Reviewed-by: Paul Durrant <paul(a)xen.org> Signed-off-by: Juergen Gross <jgross(a)suse.com> Signed-off-by: Zhengchao Shao <shaozhengchao(a)huawei.com> Reviewed-by: Yue Haibing <yuehaibing(a)huawei.com> Reviewed-by: Xiu Jianfeng <xiujianfeng(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- drivers/net/xen-netback/netback.c | 223 ++++++++++++++++-------------- 1 file changed, 123 insertions(+), 100 deletions(-) diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c index 41bdfb684d46..1036d903038c 100644 --- a/drivers/net/xen-netback/netback.c +++ b/drivers/net/xen-netback/netback.c @@ -323,10 +323,13 @@ static int xenvif_count_requests(struct xenvif_queue *queue, struct xenvif_tx_cb { - u16 pending_idx; + u16 copy_pending_idx[XEN_NETBK_LEGACY_SLOTS_MAX + 1]; + u8 copy_count; }; #define XENVIF_TX_CB(skb) ((struct xenvif_tx_cb *)(skb)->cb) +#define copy_pending_idx(skb, i) (XENVIF_TX_CB(skb)->copy_pending_idx[i]) +#define copy_count(skb) (XENVIF_TX_CB(skb)->copy_count) static inline void xenvif_tx_create_map_op(struct xenvif_queue *queue, u16 pending_idx, @@ -361,31 +364,93 @@ static inline struct sk_buff *xenvif_alloc_skb(unsigned int size) return skb; } -static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif_queue *queue, - struct sk_buff *skb, - struct xen_netif_tx_request *txp, - struct gnttab_map_grant_ref *gop, - unsigned int frag_overflow, - struct sk_buff *nskb) +static void xenvif_get_requests(struct xenvif_queue *queue, + struct sk_buff *skb, + struct xen_netif_tx_request *first, + struct xen_netif_tx_request *txfrags, + unsigned *copy_ops, + unsigned *map_ops, + unsigned int frag_overflow, + struct sk_buff *nskb, + unsigned int extra_count, + unsigned int data_len) { struct skb_shared_info *shinfo = skb_shinfo(skb); skb_frag_t *frags = shinfo->frags; - u16 pending_idx = XENVIF_TX_CB(skb)->pending_idx; - int start; + u16 pending_idx; pending_ring_idx_t index; unsigned int nr_slots; + struct gnttab_copy *cop = queue->tx_copy_ops + *copy_ops; + struct gnttab_map_grant_ref *gop = queue->tx_map_ops + *map_ops; + struct xen_netif_tx_request *txp = first; + + nr_slots = shinfo->nr_frags + 1; + + copy_count(skb) = 0; + + /* Create copy ops for exactly data_len bytes into the skb head. */ + __skb_put(skb, data_len); + while (data_len > 0) { + int amount = data_len > txp->size ? txp->size : data_len; + + cop->source.u.ref = txp->gref; + cop->source.domid = queue->vif->domid; + cop->source.offset = txp->offset; + + cop->dest.domid = DOMID_SELF; + cop->dest.offset = (offset_in_page(skb->data + + skb_headlen(skb) - + data_len)) & ~XEN_PAGE_MASK; + cop->dest.u.gmfn = virt_to_gfn(skb->data + skb_headlen(skb) + - data_len); + + cop->len = amount; + cop->flags = GNTCOPY_source_gref; - nr_slots = shinfo->nr_frags; + index = pending_index(queue->pending_cons); + pending_idx = queue->pending_ring[index]; + callback_param(queue, pending_idx).ctx = NULL; + copy_pending_idx(skb, copy_count(skb)) = pending_idx; + copy_count(skb)++; + + cop++; + data_len -= amount; - /* Skip first skb fragment if it is on same page as header fragment. */ - start = (frag_get_pending_idx(&shinfo->frags[0]) == pending_idx); + if (amount == txp->size) { + /* The copy op covered the full tx_request */ + + memcpy(&queue->pending_tx_info[pending_idx].req, + txp, sizeof(*txp)); + queue->pending_tx_info[pending_idx].extra_count = + (txp == first) ? extra_count : 0; + + if (txp == first) + txp = txfrags; + else + txp++; + queue->pending_cons++; + nr_slots--; + } else { + /* The copy op partially covered the tx_request. + * The remainder will be mapped. + */ + txp->offset += amount; + txp->size -= amount; + } + } - for (shinfo->nr_frags = start; shinfo->nr_frags < nr_slots; - shinfo->nr_frags++, txp++, gop++) { + for (shinfo->nr_frags = 0; shinfo->nr_frags < nr_slots; + shinfo->nr_frags++, gop++) { index = pending_index(queue->pending_cons++); pending_idx = queue->pending_ring[index]; - xenvif_tx_create_map_op(queue, pending_idx, txp, 0, gop); + xenvif_tx_create_map_op(queue, pending_idx, txp, + txp == first ? extra_count : 0, gop); frag_set_pending_idx(&frags[shinfo->nr_frags], pending_idx); + + if (txp == first) + txp = txfrags; + else + txp++; } if (frag_overflow) { @@ -406,7 +471,8 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif_queue *que skb_shinfo(skb)->frag_list = nskb; } - return gop; + (*copy_ops) = cop - queue->tx_copy_ops; + (*map_ops) = gop - queue->tx_map_ops; } static inline void xenvif_grant_handle_set(struct xenvif_queue *queue, @@ -442,7 +508,7 @@ static int xenvif_tx_check_gop(struct xenvif_queue *queue, struct gnttab_copy **gopp_copy) { struct gnttab_map_grant_ref *gop_map = *gopp_map; - u16 pending_idx = XENVIF_TX_CB(skb)->pending_idx; + u16 pending_idx; /* This always points to the shinfo of the skb being checked, which * could be either the first or the one on the frag_list */ @@ -453,24 +519,37 @@ static int xenvif_tx_check_gop(struct xenvif_queue *queue, struct skb_shared_info *first_shinfo = NULL; int nr_frags = shinfo->nr_frags; const bool sharedslot = nr_frags && - frag_get_pending_idx(&shinfo->frags[0]) == pending_idx; + frag_get_pending_idx(&shinfo->frags[0]) == + copy_pending_idx(skb, copy_count(skb) - 1); int i, err; - /* Check status of header. */ - err = (*gopp_copy)->status; - if (unlikely(err)) { - if (net_ratelimit()) - netdev_dbg(queue->vif->dev, - "Grant copy of header failed! status: %d pending_idx: %u ref: %u\n", - (*gopp_copy)->status, - pending_idx, - (*gopp_copy)->source.u.ref); - /* The first frag might still have this slot mapped */ - if (!sharedslot) - xenvif_idx_release(queue, pending_idx, - XEN_NETIF_RSP_ERROR); + for (i = 0; i < copy_count(skb); i++) { + int newerr; + + /* Check status of header. */ + pending_idx = copy_pending_idx(skb, i); + + newerr = (*gopp_copy)->status; + if (likely(!newerr)) { + /* The first frag might still have this slot mapped */ + if (i < copy_count(skb) - 1 || !sharedslot) + xenvif_idx_release(queue, pending_idx, + XEN_NETIF_RSP_OKAY); + } else { + err = newerr; + if (net_ratelimit()) + netdev_dbg(queue->vif->dev, + "Grant copy of header failed! status: %d pending_idx: %u ref: %u\n", + (*gopp_copy)->status, + pending_idx, + (*gopp_copy)->source.u.ref); + /* The first frag might still have this slot mapped */ + if (i < copy_count(skb) - 1 || !sharedslot) + xenvif_idx_release(queue, pending_idx, + XEN_NETIF_RSP_ERROR); + } + (*gopp_copy)++; } - (*gopp_copy)++; check_frags: for (i = 0; i < nr_frags; i++, gop_map++) { @@ -517,14 +596,6 @@ static int xenvif_tx_check_gop(struct xenvif_queue *queue, if (err) continue; - /* First error: if the header haven't shared a slot with the - * first frag, release it as well. - */ - if (!sharedslot) - xenvif_idx_release(queue, - XENVIF_TX_CB(skb)->pending_idx, - XEN_NETIF_RSP_OKAY); - /* Invalidate preceding fragments of this skb. */ for (j = 0; j < i; j++) { pending_idx = frag_get_pending_idx(&shinfo->frags[j]); @@ -794,7 +865,6 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue, unsigned *copy_ops, unsigned *map_ops) { - struct gnttab_map_grant_ref *gop = queue->tx_map_ops; struct sk_buff *skb, *nskb; int ret; unsigned int frag_overflow; @@ -876,8 +946,12 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue, continue; } + data_len = (txreq.size > XEN_NETBACK_TX_COPY_LEN) ? + XEN_NETBACK_TX_COPY_LEN : txreq.size; + ret = xenvif_count_requests(queue, &txreq, extra_count, txfrags, work_to_do); + if (unlikely(ret < 0)) break; @@ -903,9 +977,8 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue, index = pending_index(queue->pending_cons); pending_idx = queue->pending_ring[index]; - data_len = (txreq.size > XEN_NETBACK_TX_COPY_LEN && - ret < XEN_NETBK_LEGACY_SLOTS_MAX) ? - XEN_NETBACK_TX_COPY_LEN : txreq.size; + if (ret >= XEN_NETBK_LEGACY_SLOTS_MAX - 1 && data_len < txreq.size) + data_len = txreq.size; skb = xenvif_alloc_skb(data_len); if (unlikely(skb == NULL)) { @@ -916,8 +989,6 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue, } skb_shinfo(skb)->nr_frags = ret; - if (data_len < txreq.size) - skb_shinfo(skb)->nr_frags++; /* At this point shinfo->nr_frags is in fact the number of * slots, which can be as large as XEN_NETBK_LEGACY_SLOTS_MAX. */ @@ -979,54 +1050,19 @@ static void xenvif_tx_build_gops(struct xenvif_queue *queue, type); } - XENVIF_TX_CB(skb)->pending_idx = pending_idx; - - __skb_put(skb, data_len); - queue->tx_copy_ops[*copy_ops].source.u.ref = txreq.gref; - queue->tx_copy_ops[*copy_ops].source.domid = queue->vif->domid; - queue->tx_copy_ops[*copy_ops].source.offset = txreq.offset; - - queue->tx_copy_ops[*copy_ops].dest.u.gmfn = - virt_to_gfn(skb->data); - queue->tx_copy_ops[*copy_ops].dest.domid = DOMID_SELF; - queue->tx_copy_ops[*copy_ops].dest.offset = - offset_in_page(skb->data) & ~XEN_PAGE_MASK; - - queue->tx_copy_ops[*copy_ops].len = data_len; - queue->tx_copy_ops[*copy_ops].flags = GNTCOPY_source_gref; - - (*copy_ops)++; - - if (data_len < txreq.size) { - frag_set_pending_idx(&skb_shinfo(skb)->frags[0], - pending_idx); - xenvif_tx_create_map_op(queue, pending_idx, &txreq, - extra_count, gop); - gop++; - } else { - frag_set_pending_idx(&skb_shinfo(skb)->frags[0], - INVALID_PENDING_IDX); - memcpy(&queue->pending_tx_info[pending_idx].req, - &txreq, sizeof(txreq)); - queue->pending_tx_info[pending_idx].extra_count = - extra_count; - } - - queue->pending_cons++; - - gop = xenvif_get_requests(queue, skb, txfrags, gop, - frag_overflow, nskb); + xenvif_get_requests(queue, skb, &txreq, txfrags, copy_ops, + map_ops, frag_overflow, nskb, extra_count, + data_len); __skb_queue_tail(&queue->tx_queue, skb); queue->tx.req_cons = idx; - if (((gop-queue->tx_map_ops) >= ARRAY_SIZE(queue->tx_map_ops)) || + if ((*map_ops >= ARRAY_SIZE(queue->tx_map_ops)) || (*copy_ops >= ARRAY_SIZE(queue->tx_copy_ops))) break; } - (*map_ops) = gop - queue->tx_map_ops; return; } @@ -1105,9 +1141,8 @@ static int xenvif_tx_submit(struct xenvif_queue *queue) while ((skb = __skb_dequeue(&queue->tx_queue)) != NULL) { struct xen_netif_tx_request *txp; u16 pending_idx; - unsigned data_len; - pending_idx = XENVIF_TX_CB(skb)->pending_idx; + pending_idx = copy_pending_idx(skb, 0); txp = &queue->pending_tx_info[pending_idx].req; /* Check the remap error code. */ @@ -1126,18 +1161,6 @@ static int xenvif_tx_submit(struct xenvif_queue *queue) continue; } - data_len = skb->len; - callback_param(queue, pending_idx).ctx = NULL; - if (data_len < txp->size) { - /* Append the packet payload as a fragment. */ - txp->offset += data_len; - txp->size -= data_len; - } else { - /* Schedule a response immediately. */ - xenvif_idx_release(queue, pending_idx, - XEN_NETIF_RSP_OKAY); - } - if (txp->flags & XEN_NETTXF_csum_blank) skb->ip_summed = CHECKSUM_PARTIAL; else if (txp->flags & XEN_NETTXF_data_validated) @@ -1314,7 +1337,7 @@ static inline void xenvif_tx_dealloc_action(struct xenvif_queue *queue) /* Called after netfront has transmitted */ int xenvif_tx_action(struct xenvif_queue *queue, int budget) { - unsigned nr_mops, nr_cops = 0; + unsigned nr_mops = 0, nr_cops = 0; int work_done, ret; if (unlikely(!tx_work_todo(queue))) -- 2.25.1

1 2

[PATCH openEuler-1.0-LTS] arm64: fix a concurrency issue in emulation_proc_handler()
by Yongqiang Liu 16 Dec '22

16 Dec '22

From: ruanjinjie <ruanjinjie(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I65T0J CVE: NA ------------------------------- In emulation_proc_handler(), read and write operations are performed on insn->current_mode. In the concurrency scenario, mutex only protects writing insn->current_mode, and not protects the read. Suppose there are two concurrent tasks, task1 updates insn->current_mode to INSN_EMULATE in the critical section, the prev_mode of task2 is still the old data INSN_UNDEF of insn->current_mode. As a result, two tasks call update_insn_emulation_mode twice with prev_mode = INSN_UNDEF and current_mode = INSN_EMULATE, then call register_emulation_hooks twice, resulting in a list_add double problem. Call trace: __list_add_valid+0xd8/0xe4 register_undef_hook+0x94/0x13c update_insn_emulation_mode+0xd0/0x12c emulation_proc_handler+0xd8/0xf4 proc_sys_call_handler+0x140/0x250 proc_sys_write+0x1c/0x2c new_sync_write+0xec/0x18c vfs_write+0x214/0x2ac ksys_write+0x70/0xfc __arm64_sys_write+0x24/0x30 el0_svc_common.constprop.0+0x7c/0x1bc do_el0_svc+0x2c/0x94 el0_svc+0x20/0x30 el0_sync_handler+0xb0/0xb4 el0_sync+0x160/0x180 Fixes: 08f3f0b2fe7e ("arm64: fix oops in concurrently setting insn_emulation sysctls") Signed-off-by: ruanjinjie <ruanjinjie(a)huawei.com> Reviewed-by: Zhang Jianhua <chris.zjh(a)huawei.com> Reviewed-by: Liao Chang <liaochang1(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- arch/arm64/kernel/armv8_deprecated.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kernel/armv8_deprecated.c b/arch/arm64/kernel/armv8_deprecated.c index 88d3454f2848..07a0095956f1 100644 --- a/arch/arm64/kernel/armv8_deprecated.c +++ b/arch/arm64/kernel/armv8_deprecated.c @@ -211,10 +211,12 @@ static int emulation_proc_handler(struct ctl_table *table, int write, loff_t *ppos) { int ret = 0; - struct insn_emulation *insn = container_of(table->data, struct insn_emulation, current_mode); - enum insn_emulation_mode prev_mode = insn->current_mode; + struct insn_emulation *insn; + enum insn_emulation_mode prev_mode; mutex_lock(&insn_emulation_mutex); + insn = container_of(table->data, struct insn_emulation, current_mode); + prev_mode = insn->current_mode; ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); if (ret || !write || prev_mode == insn->current_mode) -- 2.25.1

1 0

[PATCH openEuler-1.0-LTS 1/2] sched/qos: Don't unthrottle cfs_rq when cfs_rq is throttled by qos
by Yongqiang Liu 16 Dec '22

16 Dec '22

From: Zhang Qiao <zhangqiao22(a)huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I64OUS CVE: NA ------------------------------- When a cfs_rq throttled by qos, mark cfs_rq->throttled as 1, and cfs bw will unthrottled this cfs_rq by mistake, it cause a list_del_valid warning. So add macro QOS_THROTTLED(=2), when a cfs_rq is throttled by qos, we mark the cfs_rq->throttled as QOS_THROTTLED, will check the value of cfs_rq->throttled before unthrottle a cfs_rq. Signed-off-by: Zhang Qiao <zhangqiao22(a)huawei.com> Reviewed-by: Chen Hui <judy.chenhui(a)huawei.com> Reviewed-by: zheng zucheng <zhengzucheng(a)huawei.com> Signed-off-by: Yongqiang Liu <liuyongqiang13(a)huawei.com> --- kernel/sched/fair.c | 59 ++++++++++++++++++++++++++------------------- 1 file changed, 34 insertions(+), 25 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index a34ca843bf0a..0dba06ce0677 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -103,6 +103,13 @@ int __weak arch_asym_cpu_priority(int cpu) #endif #ifdef CONFIG_QOS_SCHED + +/* + * To distinguish cfs bw, use QOS_THROTTLED mark cfs_rq->throttled + * when qos throttled(and cfs bw throttle mark cfs_rq->throttled as 1). + */ +#define QOS_THROTTLED 2 + static DEFINE_PER_CPU_SHARED_ALIGNED(struct list_head, qos_throttled_cfs_rq); static DEFINE_PER_CPU_SHARED_ALIGNED(struct hrtimer, qos_overload_timer); static DEFINE_PER_CPU(int, qos_cpu_overload); @@ -4649,6 +4656,14 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) se = cfs_rq->tg->se[cpu_of(rq)]; +#ifdef CONFIG_QOS_SCHED + /* + * if this cfs_rq throttled by qos, not need unthrottle it. + */ + if (cfs_rq->throttled == QOS_THROTTLED) + return; +#endif + cfs_rq->throttled = 0; update_rq_clock(rq); @@ -7066,37 +7081,18 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ #ifdef CONFIG_QOS_SCHED static void start_qos_hrtimer(int cpu); -static int qos_tg_unthrottle_up(struct task_group *tg, void *data) -{ - struct rq *rq = data; - struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; - - cfs_rq->throttle_count--; - - return 0; -} - -static int qos_tg_throttle_down(struct task_group *tg, void *data) -{ - struct rq *rq = data; - struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; - - cfs_rq->throttle_count++; - - return 0; -} - static void throttle_qos_cfs_rq(struct cfs_rq *cfs_rq) { struct rq *rq = rq_of(cfs_rq); struct sched_entity *se; long task_delta, idle_task_delta, dequeue = 1; + unsigned int prev_nr = rq->cfs.h_nr_running; se = cfs_rq->tg->se[cpu_of(rq_of(cfs_rq))]; /* freeze hierarchy runnable averages while throttled */ rcu_read_lock(); - walk_tg_tree_from(cfs_rq->tg, qos_tg_throttle_down, tg_nop, (void *)rq); + walk_tg_tree_from(cfs_rq->tg, tg_throttle_down, tg_nop, (void *)rq); rcu_read_unlock(); task_delta = cfs_rq->h_nr_running; @@ -7118,12 +7114,14 @@ static void throttle_qos_cfs_rq(struct cfs_rq *cfs_rq) if (!se) { sub_nr_running(rq, task_delta); + if (prev_nr >= 2 && prev_nr - task_delta < 2) + overload_clear(rq); } if (list_empty(&per_cpu(qos_throttled_cfs_rq, cpu_of(rq)))) start_qos_hrtimer(cpu_of(rq)); - cfs_rq->throttled = 1; + cfs_rq->throttled = QOS_THROTTLED; list_add(&cfs_rq->qos_throttled_list, &per_cpu(qos_throttled_cfs_rq, cpu_of(rq))); @@ -7135,8 +7133,11 @@ static void unthrottle_qos_cfs_rq(struct cfs_rq *cfs_rq) struct sched_entity *se; int enqueue = 1; long task_delta, idle_task_delta; + unsigned int prev_nr = rq->cfs.h_nr_running; se = cfs_rq->tg->se[cpu_of(rq)]; + if (cfs_rq->throttled != QOS_THROTTLED) + return; cfs_rq->throttled = 0; @@ -7145,7 +7146,7 @@ static void unthrottle_qos_cfs_rq(struct cfs_rq *cfs_rq) /* update hierarchical throttle state */ rcu_read_lock(); - walk_tg_tree_from(cfs_rq->tg, tg_nop, qos_tg_unthrottle_up, (void *)rq); + walk_tg_tree_from(cfs_rq->tg, tg_nop, tg_unthrottle_up, (void *)rq); rcu_read_unlock(); if (!cfs_rq->load.weight) @@ -7167,12 +7168,20 @@ static void unthrottle_qos_cfs_rq(struct cfs_rq *cfs_rq) break; } - assert_list_leaf_cfs_rq(rq); - if (!se) { add_nr_running(rq, task_delta); + if (prev_nr < 2 && prev_nr + task_delta >= 2) + overload_set(rq); + } + + for_each_sched_entity(se) { + cfs_rq = cfs_rq_of(se); + + list_add_leaf_cfs_rq(cfs_rq); } + assert_list_leaf_cfs_rq(rq); + /* Determine whether we need to wake up potentially idle CPU: */ if (rq->curr == rq->idle && rq->cfs.nr_running) resched_curr(rq); -- 2.25.1

1 1