- Kernel - mailweb.openeuler.org

[PATCH 01/48] remoteproc: Fix wrong rvring index computation
by Yang Yingliang 12 May '20

12 May '20

From: Clement Leger <cleger(a)kalray.eu> commit 00a0eec59ddbb1ce966b19097d8a8d2f777e726a upstream. Index of rvring is computed using pointer arithmetic. However, since rvring->rvdev->vring is the base of the vring array, computation of rvring idx should be reversed. It previously lead to writing at negative indices in the resource table. Signed-off-by: Clement Leger <cleger(a)kalray.eu> Link: https://lore.kernel.org/r/20191004073736.8327-1-cleger@kalray.eu Signed-off-by: Bjorn Andersson <bjorn.andersson(a)linaro.org> Cc: Doug Anderson <dianders(a)chromium.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/remoteproc/remoteproc_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/remoteproc/remoteproc_core.c b/drivers/remoteproc/remoteproc_core.c index abbef17..d5ff272 100644 --- a/drivers/remoteproc/remoteproc_core.c +++ b/drivers/remoteproc/remoteproc_core.c @@ -289,7 +289,7 @@ void rproc_free_vring(struct rproc_vring *rvring) { int size = PAGE_ALIGN(vring_size(rvring->len, rvring->align)); struct rproc *rproc = rvring->rvdev->rproc; - int idx = rvring->rvdev->vring - rvring; + int idx = rvring - rvring->rvdev->vring; struct fw_rsc_vdev *rsc; dma_free_coherent(rproc->dev.parent, size, rvring->va, rvring->dma); -- 1.8.3

1 95

[PATCH 001/124] ext4: fix extent_status fragmentation for plain files
by Yang Yingliang 12 May '20

12 May '20

From: Dmitry Monakhov <dmonakhov(a)gmail.com> [ Upstream commit 4068664e3cd2312610ceac05b74c4cf1853b8325 ] Extents are cached in read_extent_tree_block(); as a result, extents are not cached for inodes with depth == 0 when we try to find the extent using ext4_find_extent(). The result of the lookup is cached in ext4_map_blocks() but is only a subset of the extent on disk. As a result, the contents of extents status cache can get very badly fragmented for certain workloads, such as a random 4k read workload. File size of /mnt/test is 33554432 (8192 blocks of 4096 bytes) ext: logical_offset: physical_offset: length: expected: flags: 0: 0.. 8191: 40960.. 49151: 8192: last,eof $ perf record -e 'ext4:ext4_es_*' /root/bin/fio --name=t --direct=0 --rw=randread --bs=4k --filesize=32M --size=32M --filename=/mnt/test $ perf script | grep ext4_es_insert_extent | head -n 10 fio 131 [000] 13.975421: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [494/1) mapped 41454 status W fio 131 [000] 13.975939: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [6064/1) mapped 47024 status W fio 131 [000] 13.976467: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [6907/1) mapped 47867 status W fio 131 [000] 13.976937: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [3850/1) mapped 44810 status W fio 131 [000] 13.977440: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [3292/1) mapped 44252 status W fio 131 [000] 13.977931: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [6882/1) mapped 47842 status W fio 131 [000] 13.978376: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [3117/1) mapped 44077 status W fio 131 [000] 13.978957: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [2896/1) mapped 43856 status W fio 131 [000] 13.979474: ext4:ext4_es_insert_extent: dev 253,0 ino 12 es [7479/1) mapped 48439 status W Fix this by caching the extents for inodes with depth == 0 in ext4_find_extent(). [ Renamed ext4_es_cache_extents() to ext4_cache_extents() since this newly added function is not in extents_cache.c, and to avoid potential visual confusion with ext4_es_cache_extent(). -TYT ] Signed-off-by: Dmitry Monakhov <dmonakhov(a)gmail.com> Link: https://lore.kernel.org/r/20191106122502.19986-1-dmonakhov@gmail.com Signed-off-by: Theodore Ts'o <tytso(a)mit.edu> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- fs/ext4/extents.c | 47 +++++++++++++++++++++++++++-------------------- 1 file changed, 27 insertions(+), 20 deletions(-) diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 90862ee..0f17c54 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -510,6 +510,30 @@ int ext4_ext_check_inode(struct inode *inode) return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode), 0); } +static void ext4_cache_extents(struct inode *inode, + struct ext4_extent_header *eh) +{ + struct ext4_extent *ex = EXT_FIRST_EXTENT(eh); + ext4_lblk_t prev = 0; + int i; + + for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) { + unsigned int status = EXTENT_STATUS_WRITTEN; + ext4_lblk_t lblk = le32_to_cpu(ex->ee_block); + int len = ext4_ext_get_actual_len(ex); + + if (prev && (prev != lblk)) + ext4_es_cache_extent(inode, prev, lblk - prev, ~0, + EXTENT_STATUS_HOLE); + + if (ext4_ext_is_unwritten(ex)) + status = EXTENT_STATUS_UNWRITTEN; + ext4_es_cache_extent(inode, lblk, len, + ext4_ext_pblock(ex), status); + prev = lblk + len; + } +} + static struct buffer_head * __read_extent_tree_block(const char *function, unsigned int line, struct inode *inode, ext4_fsblk_t pblk, int depth, @@ -544,26 +568,7 @@ int ext4_ext_check_inode(struct inode *inode) */ if (!(flags & EXT4_EX_NOCACHE) && depth == 0) { struct ext4_extent_header *eh = ext_block_hdr(bh); - struct ext4_extent *ex = EXT_FIRST_EXTENT(eh); - ext4_lblk_t prev = 0; - int i; - - for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) { - unsigned int status = EXTENT_STATUS_WRITTEN; - ext4_lblk_t lblk = le32_to_cpu(ex->ee_block); - int len = ext4_ext_get_actual_len(ex); - - if (prev && (prev != lblk)) - ext4_es_cache_extent(inode, prev, - lblk - prev, ~0, - EXTENT_STATUS_HOLE); - - if (ext4_ext_is_unwritten(ex)) - status = EXTENT_STATUS_UNWRITTEN; - ext4_es_cache_extent(inode, lblk, len, - ext4_ext_pblock(ex), status); - prev = lblk + len; - } + ext4_cache_extents(inode, eh); } return bh; errout: @@ -911,6 +916,8 @@ struct ext4_ext_path * path[0].p_bh = NULL; i = depth; + if (!(flags & EXT4_EX_NOCACHE) && depth == 0) + ext4_cache_extents(inode, eh); /* walk through the tree */ while (i) { ext_debug("depth %d: num %d, max %d\n", -- 1.8.3

1 247

[PATCH 1/2] drivers sysctl: add read and write interface of pmbus
by Yang Yingliang 11 May '20

11 May '20

From: fengsheng <fengsheng5(a)huawei.com> driver inclusion category: feature bugzilla: NA CVE: NA 1. add interface: sysctl_pmbus_write_common 2. add interface: sysctl_pmbus_read_common Signed-off-by: fengsheng <fengsheng5(a)huawei.com> Reviewed-by: wuyang <wuyang7(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/soc/hisilicon/sysctl/sysctl_drv.c | 2 +- drivers/soc/hisilicon/sysctl/sysctl_pmbus.c | 113 ++++++++++++++++++++++------ drivers/soc/hisilicon/sysctl/sysctl_pmbus.h | 16 ++++ 3 files changed, 108 insertions(+), 23 deletions(-) diff --git a/drivers/soc/hisilicon/sysctl/sysctl_drv.c b/drivers/soc/hisilicon/sysctl/sysctl_drv.c index 3899fac..c4cb157 100644 --- a/drivers/soc/hisilicon/sysctl/sysctl_drv.c +++ b/drivers/soc/hisilicon/sysctl/sysctl_drv.c @@ -48,7 +48,7 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #define DEBUG -#define SYSCTL_DRIVER_VERSION "1.9.39.1" +#define SYSCTL_DRIVER_VERSION "1.9.60.0" unsigned int g_sysctrl_debug; diff --git a/drivers/soc/hisilicon/sysctl/sysctl_pmbus.c b/drivers/soc/hisilicon/sysctl/sysctl_pmbus.c index 74d06a8..92eeba6 100644 --- a/drivers/soc/hisilicon/sysctl/sysctl_pmbus.c +++ b/drivers/soc/hisilicon/sysctl/sysctl_pmbus.c @@ -190,6 +190,9 @@ int InitPmbus(u8 chip_id) his_sysctrl_reg_wr(base, I2C_SS_SCL_HCNT_OFFSET, I2C_SS_SCLHCNT); /* ulSclLow > 1.5us */ his_sysctrl_reg_wr(base, I2C_SS_SCL_LCNT_OFFSET, I2C_SS_SCLLCNT); + /* set sda_hold_fs 1us > 250ns */ + his_sysctrl_reg_wr(base, I2C_SDA_HOLD_OFFSET, I2C_SS_SDA_HOLD_FS); + his_sysctrl_reg_wr(base, I2C_ENABLE_OFFSET, 0x1); debug_sysctrl_print("Initialize Pmbus end\n"); @@ -233,35 +236,36 @@ int sysctl_pmbus_cfg(u8 chip_id, u8 addr, u8 page, u32 slave_addr) return 0; } -int sysctl_pmbus_write(u8 chip_id, u8 addr, u32 slave_addr, u32 data_len, u32 buf) +int sysctl_pmbus_write_common(u8 chip_id, u32 slave_addr, u32 data_len, u8 *buf) { u32 i = 0; u32 temp = 0; u32 loop = 0x1000; - u32 temp_data = addr; + u32 temp_data; void __iomem *base = NULL; if ((chip_id >= CHIP_ID_NUM_MAX) || - (data_len > DATA_NUM_MAX) || - (slave_addr >= SLAVE_ADDR_MAX)) { + (slave_addr >= SLAVE_ADDR_MAX) || + (!data_len) || (data_len > PMBUS_WRITE_LEN_MAX) || + (!buf)) { pr_err("[sysctl pmbus] write param err,chipid=0x%x,data_len=0x%x,slave_addr=0x%x!\n", chip_id, data_len, slave_addr); return SYSCTL_ERR_PARAM; } + /* clear all interrupt */ base = g_sysctl_pmbus_base[chip_id]; - his_sysctrl_reg_wr(base, I2C_INTR_RAW_OFFSET, 0x3ffff); + /* send: slave_addr[7bit] + write[1bit] */ his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, (0x2 << 0x8) | slave_addr); - if (data_len != 0) { - his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, addr); - for (i = 0; i < data_len - 1; i++) - his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, 0xff & (buf >> (i * 0x8))); + /* write data */ + for (i = 0; i < data_len - 1; i++) + his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, buf[i]); - temp_data = (0xff & (buf >> (i * 0x8))); - } + /* last data should send stop */ + temp_data = buf[i]; his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, (0x4 << 0x8) | temp_data); /* poll until send done */ @@ -273,14 +277,15 @@ int sysctl_pmbus_write(u8 chip_id, u8 addr, u32 slave_addr, u32 data_len, u32 bu /* send data failed */ if (temp & I2C_TX_ABRT) { his_sysctrl_reg_rd(base, I2C_TX_ABRT_SRC_REG, &temp); - pr_err("[sysctl pmbus]write data fail, chip_id:0x%x,slave_addr:0x%x, addr:0x%x!\r\n", - chip_id, slave_addr, addr); + pr_err("[sysctl pmbus]write data fail, chip_id:0x%x,slave_addr:0x%x\r\n", + chip_id, slave_addr); his_sysctrl_reg_rd(base, I2C_CLR_TX_ABRT_REG, &temp); return SYSCTL_ERR_FAILED; } his_sysctrl_reg_rd(base, I2C_STATUS_REG, &temp); + /* send done */ if (temp & I2C_TX_FIFO_EMPTY) { his_sysctrl_reg_rd(base, I2C_TX_FIFO_DATA_NUM_REG, &temp); if (temp == 0) @@ -289,8 +294,8 @@ int sysctl_pmbus_write(u8 chip_id, u8 addr, u32 slave_addr, u32 data_len, u32 bu loop--; if (loop == 0) { - pr_err("[sysctl pmbus]write data retry fail, chip_id:0x%x,slave_addr:0x%x, addr:0x%x!\r\n", - chip_id, slave_addr, addr); + pr_err("[sysctl pmbus]write data retry fail, chip_id:0x%x,slave_addr:0x%x\r\n", + chip_id, slave_addr); return SYSCTL_ERR_FAILED; } } @@ -298,7 +303,25 @@ int sysctl_pmbus_write(u8 chip_id, u8 addr, u32 slave_addr, u32 data_len, u32 bu return SYSCTL_ERR_OK; } -static int sysctl_pmbus_read_pre(void __iomem *base, u8 addr, u32 slave_addr, u32 data_len) +int sysctl_pmbus_write(u8 chip_id, u8 addr, u32 slave_addr, u32 data_len, u32 buf) +{ +#define TMP_LEN_MAX 5 + u8 i; + u8 tmp[TMP_LEN_MAX] = {0}; + + if (data_len > DATA_NUM_MAX) { + pr_err("[sysctl pmbus] write param err,data_len=0x%x!\n", data_len); + return SYSCTL_ERR_PARAM; + } + + tmp[0] = addr; + for (i = 0; i < data_len; i++) + tmp[i + 1] = (buf >> (i * 0x8)) & 0xff; + + return sysctl_pmbus_write_common(chip_id, slave_addr, data_len + sizeof(addr), &tmp[0]); +} + +static int sysctl_pmbus_read_pre(void __iomem *base, u32 cmd_len, u8 *cmd, u32 slave_addr, u32 data_len) { u32 i = 0; u32 fifo_num = 0; @@ -309,22 +332,29 @@ static int sysctl_pmbus_read_pre(void __iomem *base, u8 addr, u32 slave_addr, u3 return SYSCTL_ERR_PARAM; } + /* clear all interrupt */ his_sysctrl_reg_wr(base, I2C_INTR_RAW_OFFSET, 0x3ffff); + /* clear rx fifo */ his_sysctrl_reg_rd(base, I2C_RXFLR_OFFSET, &fifo_num); - debug_sysctrl_print("[sysctl_pmbus_read_byte]read pmbus , read empty rx fifo num:%d\r\n", fifo_num); for (i = 0; i < fifo_num; i++) his_sysctrl_reg_rd(base, I2C_DATA_CMD_OFFSET, &temp_byte); - his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, (0x2 << 0x8) | slave_addr); - his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, addr); - his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, (0x3 << 0x8) | slave_addr); + /* send cmd */ + if (cmd_len) { + his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, (0x2 << 0x8) | slave_addr); + for (i = 0; i < cmd_len; i++) + his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, cmd[i]); + } + /* read data */ + his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, (0x3 << 0x8) | slave_addr); i = data_len; while ((i - 1) > 0) { his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, 0x100); i--; } + /* last data should send stop */ his_sysctrl_reg_wr(base, I2C_DATA_CMD_OFFSET, 0x500); return 0; @@ -366,6 +396,43 @@ static int sysctl_pmbus_wait_data(void __iomem *base, u32 data_len) return SYSCTL_ERR_OK; } +int sysctl_pmbus_read_common(u8 chip_id, struct pmbus_read_op *op) +{ + u32 ret; + u32 i = 0; + u32 temp_byte = 0; + void __iomem *base = NULL; + + if ((chip_id >= CHIP_ID_NUM_MAX) || (!op)) { + pr_err("[sysctl pmbus]read param err,chipid=0x%x!\n", chip_id); + return SYSCTL_ERR_PARAM; + } + + if ((op->slave_addr >= SLAVE_ADDR_MAX) || + (!op->data_len) || ((op->cmd_len + op->data_len) > PMBUS_READ_LEN_MAX) || + (!op->data) || ((op->cmd_len) && (!op->cmd))) { + pr_err("[sysctl pmbus]read param err,data_len=0x%x,cmd_len=0x%x,slave_addr=0x%x\n", + op->data_len, op->cmd_len, op->slave_addr); + return SYSCTL_ERR_PARAM; + } + + base = g_sysctl_pmbus_base[chip_id]; + ret = sysctl_pmbus_read_pre(base, op->cmd_len, op->cmd, op->slave_addr, op->data_len); + if (ret != SYSCTL_ERR_OK) + return ret; + + ret = sysctl_pmbus_wait_data(base, op->data_len); + if (ret != SYSCTL_ERR_OK) + return ret; + + for (i = 0; i < op->data_len; i++) { + his_sysctrl_reg_rd(base, I2C_DATA_CMD_OFFSET, &temp_byte); + op->data[i] = temp_byte & 0xff; + } + + return SYSCTL_ERR_OK; +} + int sysctl_pmbus_read(u8 chip_id, u8 addr, u32 slave_addr, u32 data_len, u32 *buf) { u32 ret; @@ -385,7 +452,7 @@ int sysctl_pmbus_read(u8 chip_id, u8 addr, u32 slave_addr, u32 data_len, u32 *bu base = g_sysctl_pmbus_base[chip_id]; - ret = sysctl_pmbus_read_pre(base, addr, slave_addr, data_len); + ret = sysctl_pmbus_read_pre(base, sizeof(addr), &addr, slave_addr, data_len); if (ret != SYSCTL_ERR_OK) return ret; @@ -566,7 +633,7 @@ static int sysctl_cpu_convert_vol_to_vid(u32 vid_table, u32 value, u32 *vid) return SYSCTL_ERR_OK; } -int sysctl_cpu_voltage_adjust (u8 chip_id, u8 loop, u32 slave_addr, u32 value) +int sysctl_cpu_voltage_adjust(u8 chip_id, u8 loop, u32 slave_addr, u32 value) { u32 ret; u32 vid; @@ -647,5 +714,7 @@ void hip_sysctl_pmbus_exit(void) EXPORT_SYMBOL(sysctl_cpu_voltage_adjust); EXPORT_SYMBOL(sysctl_pmbus_write); EXPORT_SYMBOL(sysctl_pmbus_read); +EXPORT_SYMBOL(sysctl_pmbus_write_common); +EXPORT_SYMBOL(sysctl_pmbus_read_common); EXPORT_SYMBOL(InitPmbus); EXPORT_SYMBOL(DeInitPmbus); diff --git a/drivers/soc/hisilicon/sysctl/sysctl_pmbus.h b/drivers/soc/hisilicon/sysctl/sysctl_pmbus.h index e1a4742..ce1d83d 100644 --- a/drivers/soc/hisilicon/sysctl/sysctl_pmbus.h +++ b/drivers/soc/hisilicon/sysctl/sysctl_pmbus.h @@ -20,6 +20,13 @@ #define PAGE_NUM_MAX (0x6f) #define DATA_NUM_MAX (0x4) +#define I2C_FIFO_DEPTH (256) + +/* slave_addr use 1 fifo */ +#define PMBUS_READ_LEN_MAX (I2C_FIFO_DEPTH - 1) +/* slave_addr use 2 fifo */ +#define PMBUS_WRITE_LEN_MAX (I2C_FIFO_DEPTH - 2) + #define I2C_TX_ABRT (0x040) #define I2C_TX_ABRT_SRC_REG (0x0880) #define I2C_CLR_TX_ABRT_REG (0x0854) @@ -29,6 +36,7 @@ #define I2C_SS_SCLHCNT 0x3db #define I2C_SS_SCLLCNT 0x3e6 +#define I2C_SS_SDA_HOLD_FS 0xfa /* AVS_REG_GEN */ #define AVS_WR_OPEN_OFFSET 0x0004 @@ -77,6 +85,14 @@ #define STATUS_RPT_OFFSET 0x0AA4 #define STATUS_ERR_RPT_OFFSET 0x0AA8 +struct pmbus_read_op { + u32 slave_addr; + u32 cmd_len; + u32 data_len; + u8 *cmd; + u8 *data; +}; + /* Define the union pmbus_vout_mode */ typedef union { /* Define the struct bits */ -- 1.8.3

1 1

[PATCH 01/62] arm, bpf: Fix offset overflow for BPF_MEM BPF_DW
by Yang Yingliang 11 May '20

11 May '20

From: Luke Nelson <lukenels(a)cs.washington.edu> commit 4178417cc5359c329790a4a8f4a6604612338cca upstream. This patch fixes an incorrect check in how immediate memory offsets are computed for BPF_DW on arm. For BPF_LDX/ST/STX + BPF_DW, the 32-bit arm JIT breaks down an 8-byte access into two separate 4-byte accesses using off+0 and off+4. If off fits in imm12, the JIT emits a ldr/str instruction with the immediate and avoids the use of a temporary register. While the current check off <= 0xfff ensures that the first immediate off+0 doesn't overflow imm12, it's not sufficient for the second immediate off+4, which may cause the second access of BPF_DW to read/write the wrong address. This patch fixes the problem by changing the check to off <= 0xfff - 4 for BPF_DW, ensuring off+4 will never overflow. A side effect of simplifying the check is that it now allows using negative immediate offsets in ldr/str. This means that small negative offsets can also avoid the use of a temporary register. This patch introduces no new failures in test_verifier or test_bpf.c. Fixes: c5eae692571d6 ("ARM: net: bpf: improve 64-bit store implementation") Fixes: ec19e02b343db ("ARM: net: bpf: fix LDX instructions") Co-developed-by: Xi Wang <xi.wang(a)gmail.com> Signed-off-by: Xi Wang <xi.wang(a)gmail.com> Signed-off-by: Luke Nelson <luke.r.nels(a)gmail.com> Signed-off-by: Daniel Borkmann <daniel(a)iogearbox.net> Link: https://lore.kernel.org/bpf/20200409221752.28448-1-luke.r.nels@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- arch/arm/net/bpf_jit_32.c | 40 ++++++++++++++++++++++++---------------- 1 file changed, 24 insertions(+), 16 deletions(-) diff --git a/arch/arm/net/bpf_jit_32.c b/arch/arm/net/bpf_jit_32.c index 25b3ee8..d06293a 100644 --- a/arch/arm/net/bpf_jit_32.c +++ b/arch/arm/net/bpf_jit_32.c @@ -993,21 +993,35 @@ static inline void emit_a32_mul_r64(const s8 dst[], const s8 src[], arm_bpf_put_reg32(dst_hi, rd[0], ctx); } +static bool is_ldst_imm(s16 off, const u8 size) +{ + s16 off_max = 0; + + switch (size) { + case BPF_B: + case BPF_W: + off_max = 0xfff; + break; + case BPF_H: + off_max = 0xff; + break; + case BPF_DW: + /* Need to make sure off+4 does not overflow. */ + off_max = 0xfff - 4; + break; + } + return -off_max <= off && off <= off_max; +} + /* *(size *)(dst + off) = src */ static inline void emit_str_r(const s8 dst, const s8 src[], - s32 off, struct jit_ctx *ctx, const u8 sz){ + s16 off, struct jit_ctx *ctx, const u8 sz){ const s8 *tmp = bpf2a32[TMP_REG_1]; - s32 off_max; s8 rd; rd = arm_bpf_get_reg32(dst, tmp[1], ctx); - if (sz == BPF_H) - off_max = 0xff; - else - off_max = 0xfff; - - if (off < 0 || off > off_max) { + if (!is_ldst_imm(off, sz)) { emit_a32_mov_i(tmp[0], off, ctx); emit(ARM_ADD_R(tmp[0], tmp[0], rd), ctx); rd = tmp[0]; @@ -1036,18 +1050,12 @@ static inline void emit_str_r(const s8 dst, const s8 src[], /* dst = *(size*)(src + off) */ static inline void emit_ldx_r(const s8 dst[], const s8 src, - s32 off, struct jit_ctx *ctx, const u8 sz){ + s16 off, struct jit_ctx *ctx, const u8 sz){ const s8 *tmp = bpf2a32[TMP_REG_1]; const s8 *rd = is_stacked(dst_lo) ? tmp : dst; s8 rm = src; - s32 off_max; - - if (sz == BPF_H) - off_max = 0xff; - else - off_max = 0xfff; - if (off < 0 || off > off_max) { + if (!is_ldst_imm(off, sz)) { emit_a32_mov_i(tmp[0], off, ctx); emit(ARM_ADD_R(tmp[0], tmp[0], src), ctx); rm = tmp[0]; -- 1.8.3

1 61

Fwd: kfifo memory ordering analysis
by laokz 11 May '20

11 May '20

openEuler Kernel SIG：你好！内核社区(包括Linus)怀疑kfifo有潜在的BUG，见 https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/l… 我于去年底向社区提交了一份分析报告，用litmus验证了问题的可能性，梳理了所有kfifo相关函数和宏，提出了建议。但杳无音信，不知是丢了，还是因无实践 BUG而未受重视。 ARM是典型的弱内存序架构，因此想到转给你们供参考。如无兴趣请忽略:-) 此致 laokz

2 4

[PATCH] net/hinic: Fix TX timeout under ipip tunnel packet
by Yang Yingliang 09 May '20

09 May '20

From: Chiqijun <chiqijun(a)huawei.com> driver inclusion category: bugfix bugzilla: 4472 ----------------------------------------------------------------------- The current driver does not support ipip tunnel packet parsing. When the csum of the device is turned on, driver will get wrong inner headers, and causing device error. Signed-off-by: Chiqijun <chiqijun(a)huawei.com> Reviewed-by: Luoshaokai <luoshaokai(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/net/ethernet/huawei/hinic/hinic_tx.c | 31 ++++++++++++++++++++++------ 1 file changed, 25 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/huawei/hinic/hinic_tx.c b/drivers/net/ethernet/huawei/hinic/hinic_tx.c index 05291a5..c8492ae 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_tx.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_tx.c @@ -373,26 +373,45 @@ static int hinic_tx_csum(struct hinic_sq_task *task, u32 *queue_info, if (ip.v4->version == 4) { l3_type = IPV4_PKT_NO_CHKSUM_OFFLOAD; + l4_proto = ip.v4->protocol; } else if (ip.v4->version == 6) { + unsigned char *exthdr; + __be16 frag_off; + l3_type = IPV6_PKT; #ifdef HAVE_OUTER_IPV6_TUNNEL_OFFLOAD tunnel_type = TUNNEL_UDP_CSUM; #endif + exthdr = ip.hdr + sizeof(*ip.v6); + l4_proto = ip.v6->nexthdr; + l4.hdr = skb_transport_header(skb); + if (l4.hdr != exthdr) + ipv6_skip_exthdr(skb, exthdr - skb->data, + &l4_proto, &frag_off); } else { l3_type = UNKNOWN_L3TYPE; + l4_proto = IPPROTO_RAW; } hinic_task_set_outter_l3(task, l3_type, skb_network_header_len(skb)); - l4_tunnel_len = skb_inner_network_offset(skb) - - skb_transport_offset(skb); + if (l4_proto == IPPROTO_UDP || l4_proto == IPPROTO_GRE) { + l4_tunnel_len = skb_inner_network_offset(skb) - + skb_transport_offset(skb); + ip.hdr = skb_inner_network_header(skb); + l4.hdr = skb_inner_transport_header(skb); + network_hdr_len = skb_inner_network_header_len(skb); + } else { + tunnel_type = NOT_TUNNEL; + l4_tunnel_len = 0; - hinic_task_set_tunnel_l4(task, tunnel_type, l4_tunnel_len); + ip.hdr = skb_inner_network_header(skb); + l4.hdr = skb_transport_header(skb); + network_hdr_len = skb_network_header_len(skb); + } - ip.hdr = skb_inner_network_header(skb); - l4.hdr = skb_inner_transport_header(skb); - network_hdr_len = skb_inner_network_header_len(skb); + hinic_task_set_tunnel_l4(task, tunnel_type, l4_tunnel_len); } else { ip.hdr = skb_network_header(skb); l4.hdr = skb_transport_header(skb); -- 1.8.3

1 0

[PATCH 1/3] USB: core: Fix free-while-in-use bug in the USB S-Glibrary
by Yang Yingliang 09 May '20

09 May '20

From: Alan Stern <stern(a)rowland.harvard.edu> mainline inclusion from mainline-v5.7-rc3 commit 056ad39ee9253873522f6469c3364964a322912b category: bugfix bugzilla: 13690 CVE: CVE-2020-12464 ------------------------------------------------- FuzzUSB (a variant of syzkaller) found a free-while-still-in-use bug in the USB scatter-gather library: BUG: KASAN: use-after-free in atomic_read include/asm-generic/atomic-instrumented.h:26 [inline] BUG: KASAN: use-after-free in usb_hcd_unlink_urb+0x5f/0x170 drivers/usb/core/hcd.c:1607 Read of size 4 at addr ffff888065379610 by task kworker/u4:1/27 CPU: 1 PID: 27 Comm: kworker/u4:1 Not tainted 5.5.11 #2 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1ubuntu1 04/01/2014 Workqueue: scsi_tmf_2 scmd_eh_abort_handler Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0xce/0x128 lib/dump_stack.c:118 print_address_description.constprop.4+0x21/0x3c0 mm/kasan/report.c:374 __kasan_report+0x153/0x1cb mm/kasan/report.c:506 kasan_report+0x12/0x20 mm/kasan/common.c:639 check_memory_region_inline mm/kasan/generic.c:185 [inline] check_memory_region+0x152/0x1b0 mm/kasan/generic.c:192 __kasan_check_read+0x11/0x20 mm/kasan/common.c:95 atomic_read include/asm-generic/atomic-instrumented.h:26 [inline] usb_hcd_unlink_urb+0x5f/0x170 drivers/usb/core/hcd.c:1607 usb_unlink_urb+0x72/0xb0 drivers/usb/core/urb.c:657 usb_sg_cancel+0x14e/0x290 drivers/usb/core/message.c:602 usb_stor_stop_transport+0x5e/0xa0 drivers/usb/storage/transport.c:937 This bug occurs when cancellation of the S-G transfer races with transfer completion. When that happens, usb_sg_cancel() may continue to access the transfer's URBs after usb_sg_wait() has freed them. The bug is caused by the fact that usb_sg_cancel() does not take any sort of reference to the transfer, and so there is nothing to prevent the URBs from being deallocated while the routine is trying to use them. The fix is to take such a reference by incrementing the transfer's io->count field while the cancellation is in progres and decrementing it afterward. The transfer's URBs are not deallocated until io->complete is triggered, which happens when io->count reaches zero. Signed-off-by: Alan Stern <stern(a)rowland.harvard.edu> Reported-and-tested-by: Kyungtae Kim <kt0755(a)gmail.com> CC: <stable(a)vger.kernel.org> Link: https://lore.kernel.org/r/Pine.LNX.4.44L0.2003281615140.14837-100000@netrid… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/usb/core/message.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index 0d3fd20..fcf84bf 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -588,12 +588,13 @@ void usb_sg_cancel(struct usb_sg_request *io) int i, retval; spin_lock_irqsave(&io->lock, flags); - if (io->status) { + if (io->status || io->count == 0) { spin_unlock_irqrestore(&io->lock, flags); return; } /* shut everything down */ io->status = -ECONNRESET; + io->count++; /* Keep the request alive until we're done */ spin_unlock_irqrestore(&io->lock, flags); for (i = io->entries - 1; i >= 0; --i) { @@ -607,6 +608,12 @@ void usb_sg_cancel(struct usb_sg_request *io) dev_warn(&io->dev->dev, "%s, unlink --> %d\n", __func__, retval); } + + spin_lock_irqsave(&io->lock, flags); + io->count--; + if (!io->count) + complete(&io->complete); + spin_unlock_irqrestore(&io->lock, flags); } EXPORT_SYMBOL_GPL(usb_sg_cancel); -- 1.8.3

1 2

[PATCH 1/2] net: hns3: update hns3 version to 1.9.37.9
by Yang Yingliang 09 May '20

09 May '20

From: Shengzui You <youshengzui(a)huawei.com> driver inclusion category: other bugzilla: NA CVE: NA --------------------------------- This patch is used to modify the hns3 driver version to 1.9.37.9 Signed-off-by: Shengzui You <youshengzui(a)huawei.com> Reviewed-by: Weiwei Deng <dengweiwei(a)huawei.com> Reviewed-by: Zhaohui Zhong <zhongzhaohui(a)huawei.com> Reviewed-by: Junxin Chen <chenjunxin1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/net/ethernet/hisilicon/hns3/hnae3.h | 2 +- drivers/net/ethernet/hisilicon/hns3/hns3_cae/hns3_cae_version.h | 2 +- drivers/net/ethernet/hisilicon/hns3/hns3_enet.h | 2 +- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h | 2 +- drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h index 5253374..14b3991 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h +++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h @@ -30,7 +30,7 @@ #include <linux/pci.h> #include <linux/types.h> -#define HNAE3_MOD_VERSION "1.9.37.8" +#define HNAE3_MOD_VERSION "1.9.37.9" #define HNAE3_MIN_VECTOR_NUM 2 /* first one for misc, another for IO */ diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_cae/hns3_cae_version.h b/drivers/net/ethernet/hisilicon/hns3/hns3_cae/hns3_cae_version.h index 34ff097..b89b110 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_cae/hns3_cae_version.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_cae/hns3_cae_version.h @@ -4,7 +4,7 @@ #ifndef __HNS3_CAE_VERSION_H__ #define __HNS3_CAE_VERSION_H__ -#define HNS3_CAE_MOD_VERSION "1.9.37.8" +#define HNS3_CAE_MOD_VERSION "1.9.37.9" #define CMT_ID_LEN 8 #define RESV_LEN 3 diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h index 4567c2c..62f34e9 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h @@ -8,7 +8,7 @@ #include "hnae3.h" -#define HNS3_MOD_VERSION "1.9.37.8" +#define HNS3_MOD_VERSION "1.9.37.9" extern char hns3_driver_version[]; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h index 80270e4..15af034 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h @@ -12,7 +12,7 @@ #include "hclge_cmd.h" #include "hnae3.h" -#define HCLGE_MOD_VERSION "1.9.37.8" +#define HCLGE_MOD_VERSION "1.9.37.9" #define HCLGE_DRIVER_NAME "hclge" #define HCLGE_MAX_PF_NUM 8 diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h index bd24595..f89ee1b 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h @@ -10,7 +10,7 @@ #include "hclgevf_cmd.h" #include "hnae3.h" -#define HCLGEVF_MOD_VERSION "1.9.37.8" +#define HCLGEVF_MOD_VERSION "1.9.37.9" #define HCLGEVF_DRIVER_NAME "hclgevf" #define HCLGEVF_MAX_VLAN_ID 4095 -- 1.8.3

1 1

[PATCH] Revert "scsi: fix failing unload of a LLDD module"
by Yang Yingliang 09 May '20

09 May '20

From: Ye Bin <yebin10(a)huawei.com> hulk inclusion category: bugfix bugzilla: 34271 CVE: NA -------------------------------- This reverts commit fd72360bc94ad304136beb56e8ff2ec089113bb8. test setp: ... rmmod hisi_sas_v3_hw lsmod fdisk -l insomd hisi_sas_v3_hw.ko lsmod fdisk -l .... We get follow error when we test by above test steps. [ 3660.259153] [ffff00000116f000] pgd=00002027ffffe003, pud=00002027ffffd003, pmd=00002027cdf28003, pte=0000000000000000 [ 3660.269719] Internal error: Oops: 96000007 [#1] PREEMPT SMP [ 3660.275266] Modules linked in: hisi_sas_v3_hw(+) hisi_sas_main hns_roce_hw_v2(O) hns_roce(O) rpcrdma ib_isert iscsi_target_mod ib_iser libiscsi scsi_transport_iscsi ib_ipoib ib_umad realtek hns3(O) hclge(O) hnae3(O) crc32_ce crct10dif_ce hisi_hpre hisi_zip hisi_qm uacce hisi_trng_v2 rng_core sfc lbc ip_tables x_tables libsas scsi_transport_sas [last unloaded: hisi_sas_main] [ 3660.308227] Process smartd (pid: 19570, stack limit = 0x000000001103634d) [ 3660.314985] CPU: 31 PID: 19570 Comm: smartd Kdump: loaded Tainted: G O 4.19.36-g32894fc #1 [ 3660.324504] Hardware name: Huawei TaiShan 200 (Model 2280)/BC82AMDD, BIOS 2280-V2 CS V3.B220.02 03/27/2020 [ 3660.334110] pstate: 60400009 (nZCv daif +PAN -UAO) [ 3660.338882] pc : scsi_device_put+0x18/0x38 [ 3660.342961] lr : scsi_disk_put+0x3c/0x58 [ 3660.346865] sp : ffff0000158a3cb0 [ 3660.350164] x29: ffff0000158a3cb0 x28: ffff8027b8111000 [ 3660.355451] x27: 00000000080a005d x26: 0000000000000000 [ 3660.360738] x25: ffff8027c6310398 x24: ffff8027cd2ec410 [ 3660.366025] x23: ffff000009811000 x22: ffff80276d274750 [ 3660.371312] x21: ffff8027abdd5000 x20: ffff8027b8110800 [ 3660.376599] x19: ffff8027abdd5000 x18: 0000000000000000 [ 3660.381886] x17: 0000000000000000 x16: 0000000000000000 [ 3660.387172] x15: 0000000000000000 x14: 0000000000000000 [ 3660.392459] x13: ffff000009996cd0 x12: ffffffffffffffff [ 3660.397746] x11: ffff000009996cc8 x10: 0000000000000000 [ 3660.403033] x9 : 0000000000000000 x8 : 0000000040000000 [ 3660.408320] x7 : ffff0000098116c8 x6 : 0000000000000000 [ 3660.413607] x5 : ffff00000820ebbc x4 : ffff7e009eb8fb20 [ 3660.418894] x3 : 0000000080400009 x2 : ffff8027ae3ec600 [ 3660.424180] x1 : 71b6030ca20bb300 x0 : ffff00000116f000 [ 3660.429467] Call trace: [ 3660.431904] scsi_device_put+0x18/0x38 [ 3660.435636] scsi_disk_put+0x3c/0x58 [ 3660.439195] sd_release+0x50/0xc0 [ 3660.442496] __blkdev_put+0x20c/0x220 [ 3660.446141] blkdev_put+0x4c/0x110 [ 3660.449527] blkdev_close+0x1c/0x28 [ 3660.453000] __fput+0x88/0x1b8 [ 3660.456042] ____fput+0xc/0x18 [ 3660.459085] task_work_run+0x94/0xb0 [ 3660.462646] do_notify_resume+0x17c/0x180 [ 3660.466637] work_pending+0x8/0x10 [ 3660.470022] Code: f9000bf3 aa0003f3 f9400000 f9404c00 (f9400000) [ 3660.476089] ---[ end trace ca1d0144f9241f71 ]--- void scsi_device_put(struct scsi_device *sdev) { module_put(sdev->host->hostt->module); ---> error code put_device(&sdev->sdev_gendev); } When access "sdev->host->hostt" occurs exception, as "sdev->host->hostt" is point to the module address space which is already removed. module_delete first check module reference count, then call module exit function. So after pass module reference count check and before call module exit, we can call scsi_device_get function successfully. As "scsi: fix failing unload of a LLDD module" lead to call scsi_device_get success during remove module. We revert this patch, "scsi: fixup kernel warning during rmmod()" already fixed previous error. Signed-off-by: Ye Bin <yebin10(a)huawei.com> Reviewed-by: Hou Tao <houtao1(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- drivers/scsi/scsi.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index 7d472c2..fc1356d 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -544,6 +544,9 @@ int scsi_report_opcode(struct scsi_device *sdev, unsigned char *buffer, * Description: Gets a reference to the scsi_device and increments the use count * of the underlying LLDD module. You must hold host_lock of the * parent Scsi_Host or already have a reference when calling this. + * + * This will fail if a device is deleted or cancelled, or when the LLD module + * is in the process of being unloaded. */ int scsi_device_get(struct scsi_device *sdev) { @@ -551,12 +554,12 @@ int scsi_device_get(struct scsi_device *sdev) goto fail; if (!get_device(&sdev->sdev_gendev)) goto fail; - /* We can fail try_module_get if we're doing SCSI operations - * from module exit (like cache flush) - */ - __module_get(sdev->host->hostt->module); + if (!try_module_get(sdev->host->hostt->module)) + goto fail_put_device; return 0; +fail_put_device: + put_device(&sdev->sdev_gendev); fail: return -ENXIO; } -- 1.8.3

1 0

[PATCH] s390/mm: fix page table upgrade vs 2ndary address mode accesses
by Yang Yingliang 09 May '20

09 May '20

From: Christian Borntraeger <borntraeger(a)de.ibm.com> mainline inclusion from mainline-v5.7-rc4 commit 316ec154810960052d4586b634156c54d0778f74 category: bugfix bugzilla: 13690 CVE: CVE-2020-11884 ------------------------------------------------- A page table upgrade in a kernel section that uses secondary address mode will mess up the kernel instructions as follows: Consider the following scenario: two threads are sharing memory. On CPU1 thread 1 does e.g. strnlen_user(). That gets to old_fs = enable_sacf_uaccess(); len = strnlen_user_srst(src, size); and " la %2,0(%1)\n" " la %3,0(%0,%1)\n" " slgr %0,%0\n" " sacf 256\n" "0: srst %3,%2\n" in strnlen_user_srst(). At that point we are in secondary space mode, control register 1 points to kernel page table and instruction fetching happens via c1, rather than usual c13. Interrupts are not disabled, for obvious reasons. On CPU2 thread 2 does MAP_FIXED mmap(), forcing the upgrade of page table from 3-level to e.g. 4-level one. We'd allocated new top-level table, set it up and now we hit this: notify = 1; spin_unlock_bh(&mm->page_table_lock); } if (notify) on_each_cpu(__crst_table_upgrade, mm, 0); OK, we need to actually change over to use of new page table and we need that to happen in all threads that are currently running. Which happens to include the thread 1. IPI is delivered and we have static void __crst_table_upgrade(void *arg) { struct mm_struct *mm = arg; if (current->active_mm == mm) set_user_asce(mm); __tlb_flush_local(); } run on CPU1. That does static inline void set_user_asce(struct mm_struct *mm) { S390_lowcore.user_asce = mm->context.asce; OK, user page table address updated... __ctl_load(S390_lowcore.user_asce, 1, 1); ... and control register 1 set to it. clear_cpu_flag(CIF_ASCE_PRIMARY); } IPI is run in home space mode, so it's fine - insns are fetched using c13, which always points to kernel page table. But as soon as we return from the interrupt, previous PSW is restored, putting CPU1 back into secondary space mode, at which point we no longer get the kernel instructions from the kernel mapping. The fix is to only fixup the control registers that are currently in use for user processes during the page table update. We must also disable interrupts in enable_sacf_uaccess to synchronize the cr and thread.mm_segment updates against the on_each-cpu. Fixes: 0aaba41b58bc ("s390: remove all code using the access register mode") Cc: stable(a)vger.kernel.org # 4.15+ Reported-by: Al Viro <viro(a)zeniv.linux.org.uk> Reviewed-by: Gerald Schaefer <gerald.schaefer(a)de.ibm.com> References: CVE-2020-11884 Signed-off-by: Christian Borntraeger <borntraeger(a)de.ibm.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> Reviewed-by: Hanjun Guo <guohanjun(a)huawei.com> Reviewed-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Yang Yingliang <yangyingliang(a)huawei.com> --- arch/s390/lib/uaccess.c | 4 ++++ arch/s390/mm/pgalloc.c | 16 ++++++++++++++-- 2 files changed, 18 insertions(+), 2 deletions(-) diff --git a/arch/s390/lib/uaccess.c b/arch/s390/lib/uaccess.c index c4f8039..0267405 100644 --- a/arch/s390/lib/uaccess.c +++ b/arch/s390/lib/uaccess.c @@ -64,10 +64,13 @@ mm_segment_t enable_sacf_uaccess(void) { mm_segment_t old_fs; unsigned long asce, cr; + unsigned long flags; old_fs = current->thread.mm_segment; if (old_fs & 1) return old_fs; + /* protect against a concurrent page table upgrade */ + local_irq_save(flags); current->thread.mm_segment |= 1; asce = S390_lowcore.kernel_asce; if (likely(old_fs == USER_DS)) { @@ -83,6 +86,7 @@ mm_segment_t enable_sacf_uaccess(void) __ctl_load(asce, 7, 7); set_cpu_flag(CIF_ASCE_SECONDARY); } + local_irq_restore(flags); return old_fs; } EXPORT_SYMBOL(enable_sacf_uaccess); diff --git a/arch/s390/mm/pgalloc.c b/arch/s390/mm/pgalloc.c index 814f265..f3bc9c9 100644 --- a/arch/s390/mm/pgalloc.c +++ b/arch/s390/mm/pgalloc.c @@ -72,8 +72,20 @@ static void __crst_table_upgrade(void *arg) { struct mm_struct *mm = arg; - if (current->active_mm == mm) - set_user_asce(mm); + /* we must change all active ASCEs to avoid the creation of new TLBs */ + if (current->active_mm == mm) { + S390_lowcore.user_asce = mm->context.asce; + if (current->thread.mm_segment == USER_DS) { + __ctl_load(S390_lowcore.user_asce, 1, 1); + /* Mark user-ASCE present in CR1 */ + clear_cpu_flag(CIF_ASCE_PRIMARY); + } + if (current->thread.mm_segment == USER_DS_SACF) { + __ctl_load(S390_lowcore.user_asce, 7, 7); + /* enable_sacf_uaccess does all or nothing */ + WARN_ON(!test_cpu_flag(CIF_ASCE_SECONDARY)); + } + } __tlb_flush_local(); } -- 1.8.3

1 0

2024

2023

2022

2021

2020

2019

Kernel