April 2025 - Linux-stable-mirror

Don't backport 757b000f7b93 ("timekeeping: Fix possible inconsistencies in _COARSE clockids")

by Thomas Gleixner

Hi! It turned out that this commit causes a regression for adjtimex() users, so it will be reverted in mainline. There is a different fix available which does not have that problem, which will be marked with a Fixes tag and will be suitable for backporting. Thanks, tglx

2 months, 4 weeks

1
0
0 0

[PATCH 1/2] accel/ivpu: Fix the NPU's DPU frequency calculation

by Maciej Falkowski

From: Andrzej Kacprowski <Andrzej.Kacprowski(a)intel.com> Fix the frequency returned to the user space by the DRM_IVPU_PARAM_CORE_CLOCK_RATE GET_PARAM IOCTL. The kernel driver returned CPU frequency for MTL and bare PLL frequency for LNL - this was inconsistent and incorrect for both platforms. With this fix the driver returns maximum frequency of the NPU data processing unit (DPU) for all HW generations. This is what user space always expected. Also do not set CPU frequency in boot params - the firmware does not use frequency passed from the driver, it was only used by the early pre-production firmware. With that we can remove CPU frequency calculation code. Show NPU frequency in FREQ_CHANGE interrupt when frequency tracking is enabled. Fixes: 8a27ad81f7d3 ("accel/ivpu: Split IP and buttress code") Cc: <stable(a)vger.kernel.org> # v6.11+ Signed-off-by: Andrzej Kacprowski <Andrzej.Kacprowski(a)intel.com> Signed-off-by: Maciej Falkowski <maciej.falkowski(a)linux.intel.com> --- drivers/accel/ivpu/ivpu_drv.c | 4 +- drivers/accel/ivpu/ivpu_fw.c | 3 +- drivers/accel/ivpu/ivpu_hw.h | 11 +-- drivers/accel/ivpu/ivpu_hw_btrs.c | 126 +++++++++++++----------------- drivers/accel/ivpu/ivpu_hw_btrs.h | 6 +- include/uapi/drm/ivpu_accel.h | 4 +- 6 files changed, 66 insertions(+), 88 deletions(-) diff --git a/drivers/accel/ivpu/ivpu_drv.c b/drivers/accel/ivpu/ivpu_drv.c index 5e3888ff1022..eff1d3ca075f 100644 --- a/drivers/accel/ivpu/ivpu_drv.c +++ b/drivers/accel/ivpu/ivpu_drv.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-only /* - * Copyright (C) 2020-2024 Intel Corporation + * Copyright (C) 2020-2025 Intel Corporation */ #include <linux/firmware.h> @@ -164,7 +164,7 @@ static int ivpu_get_param_ioctl(struct drm_device *dev, void *data, struct drm_f args->value = vdev->platform; break; case DRM_IVPU_PARAM_CORE_CLOCK_RATE: - args->value = ivpu_hw_ratio_to_freq(vdev, vdev->hw->pll.max_ratio); + args->value = ivpu_hw_dpu_max_freq_get(vdev); break; case DRM_IVPU_PARAM_NUM_CONTEXTS: args->value = ivpu_get_context_count(vdev); diff --git a/drivers/accel/ivpu/ivpu_fw.c b/drivers/accel/ivpu/ivpu_fw.c index 3799231b39e7..5e1d709c6a46 100644 --- a/drivers/accel/ivpu/ivpu_fw.c +++ b/drivers/accel/ivpu/ivpu_fw.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-only /* - * Copyright (C) 2020-2024 Intel Corporation + * Copyright (C) 2020-2025 Intel Corporation */ #include <linux/firmware.h> @@ -576,7 +576,6 @@ void ivpu_fw_boot_params_setup(struct ivpu_device *vdev, struct vpu_boot_params boot_params->magic = VPU_BOOT_PARAMS_MAGIC; boot_params->vpu_id = to_pci_dev(vdev->drm.dev)->bus->number; - boot_params->frequency = ivpu_hw_pll_freq_get(vdev); /* * This param is a debug firmware feature. It switches default clock diff --git a/drivers/accel/ivpu/ivpu_hw.h b/drivers/accel/ivpu/ivpu_hw.h index 16435f2756d0..6e67e736ef8e 100644 --- a/drivers/accel/ivpu/ivpu_hw.h +++ b/drivers/accel/ivpu/ivpu_hw.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0-only */ /* - * Copyright (C) 2020-2024 Intel Corporation + * Copyright (C) 2020-2025 Intel Corporation */ #ifndef __IVPU_HW_H__ @@ -82,9 +82,9 @@ static inline u64 ivpu_hw_range_size(const struct ivpu_addr_range *range) return range->end - range->start; } -static inline u32 ivpu_hw_ratio_to_freq(struct ivpu_device *vdev, u32 ratio) +static inline u32 ivpu_hw_dpu_max_freq_get(struct ivpu_device *vdev) { - return ivpu_hw_btrs_ratio_to_freq(vdev, ratio); + return ivpu_hw_btrs_dpu_max_freq_get(vdev); } static inline void ivpu_hw_irq_clear(struct ivpu_device *vdev) @@ -92,11 +92,6 @@ static inline void ivpu_hw_irq_clear(struct ivpu_device *vdev) ivpu_hw_ip_irq_clear(vdev); } -static inline u32 ivpu_hw_pll_freq_get(struct ivpu_device *vdev) -{ - return ivpu_hw_btrs_pll_freq_get(vdev); -} - static inline u32 ivpu_hw_profiling_freq_get(struct ivpu_device *vdev) { return vdev->hw->pll.profiling_freq; diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.c b/drivers/accel/ivpu/ivpu_hw_btrs.c index 56c56012b980..91eb8a93c15b 100644 --- a/drivers/accel/ivpu/ivpu_hw_btrs.c +++ b/drivers/accel/ivpu/ivpu_hw_btrs.c @@ -1,8 +1,10 @@ // SPDX-License-Identifier: GPL-2.0-only /* - * Copyright (C) 2020-2024 Intel Corporation + * Copyright (C) 2020-2025 Intel Corporation */ +#include <linux/units.h> + #include "ivpu_drv.h" #include "ivpu_hw.h" #include "ivpu_hw_btrs.h" @@ -28,17 +30,13 @@ #define BTRS_LNL_ALL_IRQ_MASK ((u32)-1) -#define BTRS_MTL_WP_CONFIG_1_TILE_5_3_RATIO WP_CONFIG(MTL_CONFIG_1_TILE, MTL_PLL_RATIO_5_3) -#define BTRS_MTL_WP_CONFIG_1_TILE_4_3_RATIO WP_CONFIG(MTL_CONFIG_1_TILE, MTL_PLL_RATIO_4_3) -#define BTRS_MTL_WP_CONFIG_2_TILE_5_3_RATIO WP_CONFIG(MTL_CONFIG_2_TILE, MTL_PLL_RATIO_5_3) -#define BTRS_MTL_WP_CONFIG_2_TILE_4_3_RATIO WP_CONFIG(MTL_CONFIG_2_TILE, MTL_PLL_RATIO_4_3) -#define BTRS_MTL_WP_CONFIG_0_TILE_PLL_OFF WP_CONFIG(0, 0) #define PLL_CDYN_DEFAULT 0x80 #define PLL_EPP_DEFAULT 0x80 #define PLL_CONFIG_DEFAULT 0x0 -#define PLL_SIMULATION_FREQ 10000000 -#define PLL_REF_CLK_FREQ 50000000 +#define PLL_REF_CLK_FREQ 50000000ull +#define PLL_RATIO_TO_FREQ(x) ((x) * PLL_REF_CLK_FREQ) + #define PLL_TIMEOUT_US (1500 * USEC_PER_MSEC) #define IDLE_TIMEOUT_US (5 * USEC_PER_MSEC) #define TIMEOUT_US (150 * USEC_PER_MSEC) @@ -62,6 +60,8 @@ #define DCT_ENABLE 0x1 #define DCT_DISABLE 0x0 +static u32 pll_ratio_to_dpu_freq(struct ivpu_device *vdev, u32 ratio); + int ivpu_hw_btrs_irqs_clear_with_0_mtl(struct ivpu_device *vdev) { REGB_WR32(VPU_HW_BTRS_MTL_INTERRUPT_STAT, BTRS_MTL_ALL_IRQ_MASK); @@ -156,7 +156,7 @@ static int info_init_mtl(struct ivpu_device *vdev) hw->tile_fuse = BTRS_MTL_TILE_FUSE_ENABLE_BOTH; hw->sku = BTRS_MTL_TILE_SKU_BOTH; - hw->config = BTRS_MTL_WP_CONFIG_2_TILE_4_3_RATIO; + hw->config = WP_CONFIG(MTL_CONFIG_2_TILE, MTL_PLL_RATIO_4_3); return 0; } @@ -334,8 +334,8 @@ int ivpu_hw_btrs_wp_drive(struct ivpu_device *vdev, bool enable) prepare_wp_request(vdev, &wp, enable); - ivpu_dbg(vdev, PM, "PLL workpoint request: %u Hz, config: 0x%x, epp: 0x%x, cdyn: 0x%x\n", - PLL_RATIO_TO_FREQ(wp.target), wp.cfg, wp.epp, wp.cdyn); + ivpu_dbg(vdev, PM, "PLL workpoint request: %lu MHz, config: 0x%x, epp: 0x%x, cdyn: 0x%x\n", + pll_ratio_to_dpu_freq(vdev, wp.target) / HZ_PER_MHZ, wp.cfg, wp.epp, wp.cdyn); ret = wp_request_send(vdev, &wp); if (ret) { @@ -573,6 +573,39 @@ int ivpu_hw_btrs_wait_for_idle(struct ivpu_device *vdev) return REGB_POLL_FLD(VPU_HW_BTRS_LNL_VPU_STATUS, IDLE, 0x1, IDLE_TIMEOUT_US); } +static u32 pll_config_get_mtl(struct ivpu_device *vdev) +{ + return REGB_RD32(VPU_HW_BTRS_MTL_CURRENT_PLL); +} + +static u32 pll_config_get_lnl(struct ivpu_device *vdev) +{ + return REGB_RD32(VPU_HW_BTRS_LNL_PLL_FREQ); +} + +static u32 pll_ratio_to_dpu_freq_mtl(u16 ratio) +{ + return (PLL_RATIO_TO_FREQ(ratio) * 2) / 3; +} + +static u32 pll_ratio_to_dpu_freq_lnl(u16 ratio) +{ + return PLL_RATIO_TO_FREQ(ratio) / 2; +} + +static u32 pll_ratio_to_dpu_freq(struct ivpu_device *vdev, u32 ratio) +{ + if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL) + return pll_ratio_to_dpu_freq_mtl(ratio); + else + return pll_ratio_to_dpu_freq_lnl(ratio); +} + +u32 ivpu_hw_btrs_dpu_max_freq_get(struct ivpu_device *vdev) +{ + return pll_ratio_to_dpu_freq(vdev, vdev->hw->pll.max_ratio); +} + /* Handler for IRQs from Buttress core (irqB) */ bool ivpu_hw_btrs_irq_handler_mtl(struct ivpu_device *vdev, int irq) { @@ -582,9 +615,12 @@ bool ivpu_hw_btrs_irq_handler_mtl(struct ivpu_device *vdev, int irq) if (!status) return false; - if (REG_TEST_FLD(VPU_HW_BTRS_MTL_INTERRUPT_STAT, FREQ_CHANGE, status)) - ivpu_dbg(vdev, IRQ, "FREQ_CHANGE irq: %08x", - REGB_RD32(VPU_HW_BTRS_MTL_CURRENT_PLL)); + if (REG_TEST_FLD(VPU_HW_BTRS_MTL_INTERRUPT_STAT, FREQ_CHANGE, status)) { + u32 pll = pll_config_get_mtl(vdev); + + ivpu_dbg(vdev, IRQ, "FREQ_CHANGE irq, wp %08x, %lu MHz", + pll, pll_ratio_to_dpu_freq_mtl(pll) / HZ_PER_MHZ); + } if (REG_TEST_FLD(VPU_HW_BTRS_MTL_INTERRUPT_STAT, ATS_ERR, status)) { ivpu_err(vdev, "ATS_ERR irq 0x%016llx", REGB_RD64(VPU_HW_BTRS_MTL_ATS_ERR_LOG_0)); @@ -633,8 +669,12 @@ bool ivpu_hw_btrs_irq_handler_lnl(struct ivpu_device *vdev, int irq) queue_work(system_wq, &vdev->irq_dct_work); } - if (REG_TEST_FLD(VPU_HW_BTRS_LNL_INTERRUPT_STAT, FREQ_CHANGE, status)) - ivpu_dbg(vdev, IRQ, "FREQ_CHANGE irq: %08x", REGB_RD32(VPU_HW_BTRS_LNL_PLL_FREQ)); + if (REG_TEST_FLD(VPU_HW_BTRS_LNL_INTERRUPT_STAT, FREQ_CHANGE, status)) { + u32 pll = pll_config_get_lnl(vdev); + + ivpu_dbg(vdev, IRQ, "FREQ_CHANGE irq, wp %08x, %lu MHz", + pll, pll_ratio_to_dpu_freq_lnl(pll) / HZ_PER_MHZ); + } if (REG_TEST_FLD(VPU_HW_BTRS_LNL_INTERRUPT_STAT, ATS_ERR, status)) { ivpu_err(vdev, "ATS_ERR LOG1 0x%08x ATS_ERR_LOG2 0x%08x\n", @@ -717,60 +757,6 @@ void ivpu_hw_btrs_dct_set_status(struct ivpu_device *vdev, bool enable, u32 acti REGB_WR32(VPU_HW_BTRS_LNL_PCODE_MAILBOX_STATUS, val); } -static u32 pll_ratio_to_freq_mtl(u32 ratio, u32 config) -{ - u32 pll_clock = PLL_REF_CLK_FREQ * ratio; - u32 cpu_clock; - - if ((config & 0xff) == MTL_PLL_RATIO_4_3) - cpu_clock = pll_clock * 2 / 4; - else - cpu_clock = pll_clock * 2 / 5; - - return cpu_clock; -} - -u32 ivpu_hw_btrs_ratio_to_freq(struct ivpu_device *vdev, u32 ratio) -{ - struct ivpu_hw_info *hw = vdev->hw; - - if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL) - return pll_ratio_to_freq_mtl(ratio, hw->config); - else - return PLL_RATIO_TO_FREQ(ratio); -} - -static u32 pll_freq_get_mtl(struct ivpu_device *vdev) -{ - u32 pll_curr_ratio; - - pll_curr_ratio = REGB_RD32(VPU_HW_BTRS_MTL_CURRENT_PLL); - pll_curr_ratio &= VPU_HW_BTRS_MTL_CURRENT_PLL_RATIO_MASK; - - if (!ivpu_is_silicon(vdev)) - return PLL_SIMULATION_FREQ; - - return pll_ratio_to_freq_mtl(pll_curr_ratio, vdev->hw->config); -} - -static u32 pll_freq_get_lnl(struct ivpu_device *vdev) -{ - u32 pll_curr_ratio; - - pll_curr_ratio = REGB_RD32(VPU_HW_BTRS_LNL_PLL_FREQ); - pll_curr_ratio &= VPU_HW_BTRS_LNL_PLL_FREQ_RATIO_MASK; - - return PLL_RATIO_TO_FREQ(pll_curr_ratio); -} - -u32 ivpu_hw_btrs_pll_freq_get(struct ivpu_device *vdev) -{ - if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL) - return pll_freq_get_mtl(vdev); - else - return pll_freq_get_lnl(vdev); -} - u32 ivpu_hw_btrs_telemetry_offset_get(struct ivpu_device *vdev) { if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL) diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.h b/drivers/accel/ivpu/ivpu_hw_btrs.h index 1fd71b4d4ab0..a2f0877237e8 100644 --- a/drivers/accel/ivpu/ivpu_hw_btrs.h +++ b/drivers/accel/ivpu/ivpu_hw_btrs.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0-only */ /* - * Copyright (C) 2020-2024 Intel Corporation + * Copyright (C) 2020-2025 Intel Corporation */ #ifndef __IVPU_HW_BTRS_H__ @@ -13,7 +13,6 @@ #define PLL_PROFILING_FREQ_DEFAULT 38400000 #define PLL_PROFILING_FREQ_HIGH 400000000 -#define PLL_RATIO_TO_FREQ(x) ((x) * PLL_REF_CLK_FREQ) #define DCT_DEFAULT_ACTIVE_PERCENT 15u #define DCT_PERIOD_US 35300u @@ -32,12 +31,11 @@ int ivpu_hw_btrs_ip_reset(struct ivpu_device *vdev); void ivpu_hw_btrs_profiling_freq_reg_set_lnl(struct ivpu_device *vdev); void ivpu_hw_btrs_ats_print_lnl(struct ivpu_device *vdev); void ivpu_hw_btrs_clock_relinquish_disable_lnl(struct ivpu_device *vdev); +u32 ivpu_hw_btrs_dpu_max_freq_get(struct ivpu_device *vdev); bool ivpu_hw_btrs_irq_handler_mtl(struct ivpu_device *vdev, int irq); bool ivpu_hw_btrs_irq_handler_lnl(struct ivpu_device *vdev, int irq); int ivpu_hw_btrs_dct_get_request(struct ivpu_device *vdev, bool *enable); void ivpu_hw_btrs_dct_set_status(struct ivpu_device *vdev, bool enable, u32 dct_percent); -u32 ivpu_hw_btrs_pll_freq_get(struct ivpu_device *vdev); -u32 ivpu_hw_btrs_ratio_to_freq(struct ivpu_device *vdev, u32 ratio); u32 ivpu_hw_btrs_telemetry_offset_get(struct ivpu_device *vdev); u32 ivpu_hw_btrs_telemetry_size_get(struct ivpu_device *vdev); u32 ivpu_hw_btrs_telemetry_enable_get(struct ivpu_device *vdev); diff --git a/include/uapi/drm/ivpu_accel.h b/include/uapi/drm/ivpu_accel.h index 746c43bd3eb6..2f24103f4533 100644 --- a/include/uapi/drm/ivpu_accel.h +++ b/include/uapi/drm/ivpu_accel.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0-only WITH Linux-syscall-note */ /* - * Copyright (C) 2020-2024 Intel Corporation + * Copyright (C) 2020-2025 Intel Corporation */ #ifndef __UAPI_IVPU_DRM_H__ @@ -147,7 +147,7 @@ struct drm_ivpu_param { * platform type when executing on a simulator or emulator (read-only) * * %DRM_IVPU_PARAM_CORE_CLOCK_RATE: - * Current PLL frequency (read-only) + * Maximum frequency of the NPU data processing unit clock (read-only) * * %DRM_IVPU_PARAM_NUM_CONTEXTS: * Maximum number of simultaneously existing contexts (read-only) -- 2.43.0

2 months, 4 weeks

2
1
0 0

[PATCH 5.4.y] RDMA/srpt: Support specifying the srpt_service_guid parameter

by Alok Tiwari

From: Bart Van Assche <bvanassche(a)acm.org> [ Upstream commit fdfa083549de5d50ebf7f6811f33757781e838c0 ] Make loading ib_srpt with this parameter set work. The current behavior is that setting that parameter while loading the ib_srpt kernel module triggers the following kernel crash: BUG: kernel NULL pointer dereference, address: 0000000000000000 Call Trace: <TASK> parse_one+0x18c/0x1d0 parse_args+0xe1/0x230 load_module+0x8de/0xa60 init_module_from_file+0x8b/0xd0 idempotent_init_module+0x181/0x240 __x64_sys_finit_module+0x5a/0xb0 do_syscall_64+0x5f/0xe0 entry_SYSCALL_64_after_hwframe+0x6e/0x76 Cc: LiHonggang <honggangli(a)163.com> Reported-by: LiHonggang <honggangli(a)163.com> Fixes: a42d985bd5b2 ("ib_srpt: Initial SRP Target merge for v3.3-rc1") Signed-off-by: Bart Van Assche <bvanassche(a)acm.org> Link: https://lore.kernel.org/r/20240205004207.17031-1-bvanassche@acm.org Signed-off-by: Leon Romanovsky <leon(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> [Alok: Backport to 5.4.y since the commit has already been backported to 5.15y, 5.10.y, and 4.19.y] Signed-off-by: Alok Tiwari <alok.a.tiwari(a)oracle.com> --- drivers/infiniband/ulp/srpt/ib_srpt.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c index d03a4f2e006f..f5fd8c1058ce 100644 --- a/drivers/infiniband/ulp/srpt/ib_srpt.c +++ b/drivers/infiniband/ulp/srpt/ib_srpt.c @@ -79,12 +79,16 @@ module_param(srpt_srq_size, int, 0444); MODULE_PARM_DESC(srpt_srq_size, "Shared receive queue (SRQ) size."); +static int srpt_set_u64_x(const char *buffer, const struct kernel_param *kp) +{ + return kstrtou64(buffer, 16, (u64 *)kp->arg); +} static int srpt_get_u64_x(char *buffer, const struct kernel_param *kp) { return sprintf(buffer, "0x%016llx", *(u64 *)kp->arg); } -module_param_call(srpt_service_guid, NULL, srpt_get_u64_x, &srpt_service_guid, - 0444); +module_param_call(srpt_service_guid, srpt_set_u64_x, srpt_get_u64_x, + &srpt_service_guid, 0444); MODULE_PARM_DESC(srpt_service_guid, "Using this value for ioc_guid, id_ext, and cm_listen_id instead of using the node_guid of the first HCA."); -- 2.46.0

2 months, 4 weeks

3
2
0 0

[PATCH stable] xdp: Reset bpf_redirect_info before running a xdp's BPF prog.

by Sebastian Andrzej Siewior

Ricardo reported a KASAN discovered use after free in v6.6-stable. The syzbot starts a BPF program via xdp_test_run_batch() which assigns ri->tgt_value via dev_hash_map_redirect() and the return code isn't XDP_REDIRECT it looks like nonsense. So the output in bpf_warn_invalid_xdp_action() appears once. Then the TUN driver runs another BPF program (on the same CPU) which returns XDP_REDIRECT without setting ri->tgt_value first. It invokes bpf_trace_printk() to print four characters and obtain the required return value. This is enough to get xdp_do_redirect() invoked which then accesses the pointer in tgt_value which might have been already deallocated. This problem does not affect upstream because since commit 401cb7dae8130 ("net: Reference bpf_redirect_info via task_struct on PREEMPT_RT.") the per-CPU variable is referenced via task's task_struct and exists on the stack during NAPI callback. Therefore it is cleared once before the first invocation and remains valid within the RCU section of the NAPI callback. Instead of performing the huge backport of the commit (plus its fix ups) here is an alternative version which only resets the variable in question prior invoking the BPF program. Acked-by: Toke Høiland-Jørgensen <toke(a)kernel.org> Reported-by: Ricardo Cañuelo Navarro <rcn(a)igalia.com> Closes: https://lore.kernel.org/all/20250226-20250204-kasan-slab-use-after-free-rea… Fixes: 97f91a7cf04ff ("bpf: add bpf_redirect_map helper routine") Signed-off-by: Sebastian Andrzej Siewior <bigeasy(a)linutronix.de> --- I discussed this with Toke, thread starts at https://lore.kernel.org/all/20250313183911.SPAmGLyw@linutronix.de/ The commit, which this by accident, is part of v6.11-rc1. I added the commit introducing map redirects as the origin of the problem which is v4.14-rc1. The code is a bit different there it seems to work similar. Greg, feel free to decide if this is worth a CVE. include/net/xdp.h | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/include/net/xdp.h b/include/net/xdp.h index de08c8e0d1348..b39ac83618a55 100644 --- a/include/net/xdp.h +++ b/include/net/xdp.h @@ -486,7 +486,14 @@ static __always_inline u32 bpf_prog_run_xdp(const struct bpf_prog *prog, * under local_bh_disable(), which provides the needed RCU protection * for accessing map entries. */ - u32 act = __bpf_prog_run(prog, xdp, BPF_DISPATCHER_FUNC(xdp)); + struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info); + u32 act; + + if (ri->map_id || ri->map_type) { + ri->map_id = 0; + ri->map_type = BPF_MAP_TYPE_UNSPEC; + } + act = __bpf_prog_run(prog, xdp, BPF_DISPATCHER_FUNC(xdp)); if (static_branch_unlikely(&bpf_master_redirect_enabled_key)) { if (act == XDP_TX && netif_is_bond_slave(xdp->rxq->dev)) -- 2.49.0

2 months, 4 weeks

2
3
0 0

[PATCH 5.10/5.15] scsi: ufs: bsg: Set bsg_queue to NULL after removal

by Ilia Gavrilov

From: "Ilia.Gavrilov" <Ilia.Gavrilov(a)infotecs.ru> From: Guixin Liu <kanie(a)linux.alibaba.com> commit 1e95c798d8a7f70965f0f88d4657b682ff0ec75f upstream. Currently, this does not cause any issues, but I believe it is necessary to set bsg_queue to NULL after removing it to prevent potential use-after-free (UAF) access. Signed-off-by: Guixin Liu <kanie(a)linux.alibaba.com> Link: https://lore.kernel.org/r/20241218014214.64533-3-kanie@linux.alibaba.com Reviewed-by: Avri Altman <avri.altman(a)wdc.com> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Ilia Gavrilov <Ilia.Gavrilov(a)infotecs.ru> --- Backport fix for CVE-2024-54458 drivers/scsi/ufs/ufs_bsg.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/scsi/ufs/ufs_bsg.c b/drivers/scsi/ufs/ufs_bsg.c index 05c7347eda18..a7e1b011202b 100644 --- a/drivers/scsi/ufs/ufs_bsg.c +++ b/drivers/scsi/ufs/ufs_bsg.c @@ -175,6 +175,7 @@ void ufs_bsg_remove(struct ufs_hba *hba) return; bsg_remove_queue(hba->bsg_queue); + hba->bsg_queue = NULL; device_del(bsg_dev); put_device(bsg_dev); -- 2.39.5

2 months, 4 weeks

1
0
0 0

[PATCH 5.4.y] tcp/dccp: Don't use timer_pending() in reqsk_queue_unlink().

by Nathan Gao

From: Kuniyuki Iwashima <kuniyu(a)amazon.com> [ Upstream commit e8c526f2bdf1845bedaf6a478816a3d06fa78b8f ] Martin KaFai Lau reported use-after-free [0] in reqsk_timer_handler(). """ We are seeing a use-after-free from a bpf prog attached to trace_tcp_retransmit_synack. The program passes the req->sk to the bpf_sk_storage_get_tracing kernel helper which does check for null before using it. """ The commit 83fccfc3940c ("inet: fix potential deadlock in reqsk_queue_unlink()") added timer_pending() in reqsk_queue_unlink() not to call del_timer_sync() from reqsk_timer_handler(), but it introduced a small race window. Before the timer is called, expire_timers() calls detach_timer(timer, true) to clear timer->entry.pprev and marks it as not pending. If reqsk_queue_unlink() checks timer_pending() just after expire_timers() calls detach_timer(), TCP will miss del_timer_sync(); the reqsk timer will continue running and send multiple SYN+ACKs until it expires. The reported UAF could happen if req->sk is close()d earlier than the timer expiration, which is 63s by default. The scenario would be 1. inet_csk_complete_hashdance() calls inet_csk_reqsk_queue_drop(), but del_timer_sync() is missed 2. reqsk timer is executed and scheduled again 3. req->sk is accept()ed and reqsk_put() decrements rsk_refcnt, but reqsk timer still has another one, and inet_csk_accept() does not clear req->sk for non-TFO sockets 4. sk is close()d 5. reqsk timer is executed again, and BPF touches req->sk Let's not use timer_pending() by passing the caller context to __inet_csk_reqsk_queue_drop(). Note that reqsk timer is pinned, so the issue does not happen in most use cases. [1] [0] BUG: KFENCE: use-after-free read in bpf_sk_storage_get_tracing+0x2e/0x1b0 Use-after-free read at 0x00000000a891fb3a (in kfence-#1): bpf_sk_storage_get_tracing+0x2e/0x1b0 bpf_prog_5ea3e95db6da0438_tcp_retransmit_synack+0x1d20/0x1dda bpf_trace_run2+0x4c/0xc0 tcp_rtx_synack+0xf9/0x100 reqsk_timer_handler+0xda/0x3d0 run_timer_softirq+0x292/0x8a0 irq_exit_rcu+0xf5/0x320 sysvec_apic_timer_interrupt+0x6d/0x80 asm_sysvec_apic_timer_interrupt+0x16/0x20 intel_idle_irq+0x5a/0xa0 cpuidle_enter_state+0x94/0x273 cpu_startup_entry+0x15e/0x260 start_secondary+0x8a/0x90 secondary_startup_64_no_verify+0xfa/0xfb kfence-#1: 0x00000000a72cc7b6-0x00000000d97616d9, size=2376, cache=TCPv6 allocated by task 0 on cpu 9 at 260507.901592s: sk_prot_alloc+0x35/0x140 sk_clone_lock+0x1f/0x3f0 inet_csk_clone_lock+0x15/0x160 tcp_create_openreq_child+0x1f/0x410 tcp_v6_syn_recv_sock+0x1da/0x700 tcp_check_req+0x1fb/0x510 tcp_v6_rcv+0x98b/0x1420 ipv6_list_rcv+0x2258/0x26e0 napi_complete_done+0x5b1/0x2990 mlx5e_napi_poll+0x2ae/0x8d0 net_rx_action+0x13e/0x590 irq_exit_rcu+0xf5/0x320 common_interrupt+0x80/0x90 asm_common_interrupt+0x22/0x40 cpuidle_enter_state+0xfb/0x273 cpu_startup_entry+0x15e/0x260 start_secondary+0x8a/0x90 secondary_startup_64_no_verify+0xfa/0xfb freed by task 0 on cpu 9 at 260507.927527s: rcu_core_si+0x4ff/0xf10 irq_exit_rcu+0xf5/0x320 sysvec_apic_timer_interrupt+0x6d/0x80 asm_sysvec_apic_timer_interrupt+0x16/0x20 cpuidle_enter_state+0xfb/0x273 cpu_startup_entry+0x15e/0x260 start_secondary+0x8a/0x90 secondary_startup_64_no_verify+0xfa/0xfb Fixes: 83fccfc3940c ("inet: fix potential deadlock in reqsk_queue_unlink()") Reported-by: Martin KaFai Lau <martin.lau(a)kernel.org> Closes: https://lore.kernel.org/netdev/eb6684d0-ffd9-4bdc-9196-33f690c25824@linux.d… Link: https://lore.kernel.org/netdev/b55e2ca0-42f2-4b7c-b445-6ffd87ca74a0@linux.d… [1] Signed-off-by: Kuniyuki Iwashima <kuniyu(a)amazon.com> Reviewed-by: Eric Dumazet <edumazet(a)google.com> Reviewed-by: Martin KaFai Lau <martin.lau(a)kernel.org> Link: https://patch.msgid.link/20241014223312.4254-1-kuniyu@amazon.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> [Resolved conflicts due to context difference] Signed-off-by: Nathan Gao <zcgao(a)amazon.com> --- net/ipv4/inet_connection_sock.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 6766a154ff85..e1618cfd5e78 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -709,21 +709,31 @@ static bool reqsk_queue_unlink(struct request_sock *req) found = __sk_nulls_del_node_init_rcu(req_to_sk(req)); spin_unlock(lock); } - if (timer_pending(&req->rsk_timer) && del_timer_sync(&req->rsk_timer)) - reqsk_put(req); + return found; } -bool inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req) +static bool __inet_csk_reqsk_queue_drop(struct sock *sk, + struct request_sock *req, + bool from_timer) { bool unlinked = reqsk_queue_unlink(req); + if (!from_timer && timer_delete_sync(&req->rsk_timer)) + reqsk_put(req); + if (unlinked) { reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req); reqsk_put(req); } + return unlinked; } + +bool inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req) +{ + return __inet_csk_reqsk_queue_drop(sk, req, false); +} EXPORT_SYMBOL(inet_csk_reqsk_queue_drop); void inet_csk_reqsk_queue_drop_and_put(struct sock *sk, struct request_sock *req) @@ -796,7 +806,8 @@ static void reqsk_timer_handler(struct timer_list *t) return; } drop: - inet_csk_reqsk_queue_drop_and_put(sk_listener, req); + __inet_csk_reqsk_queue_drop(sk_listener, req, true); + reqsk_put(req); } static void reqsk_queue_hash_req(struct request_sock *req, -- 2.39.5 (Apple Git-154)

2 months, 4 weeks

2
1
0 0

[PATCH] loop: Properly send KOBJ_CHANGED uevent for disk device

by Thomas Weißschuh

The wording "uncork" in the code comment indicates that it is expected that the suppressed event instances are automatically sent after unsuppressing. This is not the case, they are discarded. In effect this means that no "changed" events are emitted on the device itself by default. On the other hand each discovered partition does trigger a "changed" event on the loop device itself. Therefore no event is emitted for devices without partitions. This leads to udev missing the device creation and prompting workarounds in userspace, see the linked util-linux/losetup bug. Explicitly emit the events and drop the confusingly worded comments. Link: https://github.com/util-linux/util-linux/issues/2434 Fixes: 3448914e8cc5 ("loop: Add LOOP_CONFIGURE ioctl") Cc: stable(a)vger.kernel.org Signed-off-by: Thomas Weißschuh <thomas.weissschuh(a)linutronix.de> --- drivers/block/loop.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index c05fe27a96b64f1f1ea3868510fdd0c7f4937f55..fbc67ff29e07c15f2e3b3e225a4a37df016fe9de 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -654,8 +654,8 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev, error = 0; done: - /* enable and uncork uevent now that we are done */ dev_set_uevent_suppress(disk_to_dev(lo->lo_disk), 0); + kobject_uevent(&disk_to_dev(lo->lo_disk)->kobj, KOBJ_CHANGE); return error; out_err: @@ -1115,8 +1115,8 @@ static int loop_configure(struct loop_device *lo, blk_mode_t mode, if (partscan) clear_bit(GD_SUPPRESS_PART_SCAN, &lo->lo_disk->state); - /* enable and uncork uevent now that we are done */ dev_set_uevent_suppress(disk_to_dev(lo->lo_disk), 0); + kobject_uevent(&disk_to_dev(lo->lo_disk)->kobj, KOBJ_CHANGE); loop_global_unlock(lo, is_loop); if (partscan) --- base-commit: 4701f33a10702d5fc577c32434eb62adde0a1ae1 change-id: 20250307-loop-uevent-changed-aa3690f43e03 Best regards, -- Thomas Weißschuh <thomas.weissschuh(a)linutronix.de>

2 months, 4 weeks

2
2
0 0

[PATCH AUTOSEL 5.4 1/6] MIPS: cm: Detect CM quirks from device tree

by Sasha Levin

From: Gregory CLEMENT <gregory.clement(a)bootlin.com> [ Upstream commit e27fbe16af5cfc40639de4ced67d1a866a1953e9 ] Some information that should be retrieved at runtime for the Coherence Manager can be either absent or wrong. This patch allows checking if some of this information is available from the device tree and updates the internal variable accordingly. For now, only the compatible string associated with the broken HCI is being retrieved. Signed-off-by: Gregory CLEMENT <gregory.clement(a)bootlin.com> Signed-off-by: Thomas Bogendoerfer <tsbogend(a)alpha.franken.de> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- arch/mips/include/asm/mips-cm.h | 22 ++++++++++++++++++++++ arch/mips/kernel/mips-cm.c | 14 ++++++++++++++ 2 files changed, 36 insertions(+) diff --git a/arch/mips/include/asm/mips-cm.h b/arch/mips/include/asm/mips-cm.h index 696b40beb774f..0f31324998c0a 100644 --- a/arch/mips/include/asm/mips-cm.h +++ b/arch/mips/include/asm/mips-cm.h @@ -47,6 +47,16 @@ extern phys_addr_t __mips_cm_phys_base(void); */ extern int mips_cm_is64; +/* + * mips_cm_is_l2_hci_broken - determine if HCI is broken + * + * Some CM reports show that Hardware Cache Initialization is + * complete, but in reality it's not the case. They also incorrectly + * indicate that Hardware Cache Initialization is supported. This + * flags allows warning about this broken feature. + */ +extern bool mips_cm_is_l2_hci_broken; + /** * mips_cm_error_report - Report CM cache errors */ @@ -85,6 +95,18 @@ static inline bool mips_cm_present(void) #endif } +/** + * mips_cm_update_property - update property from the device tree + * + * Retrieve the properties from the device tree if a CM node exist and + * update the internal variable based on this. + */ +#ifdef CONFIG_MIPS_CM +extern void mips_cm_update_property(void); +#else +static void mips_cm_update_property(void) {} +#endif + /** * mips_cm_has_l2sync - determine whether an L2-only sync region is present * diff --git a/arch/mips/kernel/mips-cm.c b/arch/mips/kernel/mips-cm.c index 611ef512c0b81..159354ac9335b 100644 --- a/arch/mips/kernel/mips-cm.c +++ b/arch/mips/kernel/mips-cm.c @@ -5,6 +5,7 @@ */ #include <linux/errno.h> +#include <linux/of.h> #include <linux/percpu.h> #include <linux/spinlock.h> @@ -14,6 +15,7 @@ void __iomem *mips_gcr_base; void __iomem *mips_cm_l2sync_base; int mips_cm_is64; +bool mips_cm_is_l2_hci_broken; static char *cm2_tr[8] = { "mem", "gcr", "gic", "mmio", @@ -196,6 +198,18 @@ static void mips_cm_probe_l2sync(void) mips_cm_l2sync_base = ioremap_nocache(addr, MIPS_CM_L2SYNC_SIZE); } +void mips_cm_update_property(void) +{ + struct device_node *cm_node; + + cm_node = of_find_compatible_node(of_root, NULL, "mobileye,eyeq6-cm"); + if (!cm_node) + return; + pr_info("HCI (Hardware Cache Init for the L2 cache) in GCR_L2_RAM_CONFIG from the CM3 is broken"); + mips_cm_is_l2_hci_broken = true; + of_node_put(cm_node); +} + int mips_cm_probe(void) { phys_addr_t addr; -- 2.39.5

3 months

1
5
0 0

[PATCH AUTOSEL 5.10 1/6] MIPS: cm: Detect CM quirks from device tree

by Sasha Levin

From: Gregory CLEMENT <gregory.clement(a)bootlin.com> [ Upstream commit e27fbe16af5cfc40639de4ced67d1a866a1953e9 ] Some information that should be retrieved at runtime for the Coherence Manager can be either absent or wrong. This patch allows checking if some of this information is available from the device tree and updates the internal variable accordingly. For now, only the compatible string associated with the broken HCI is being retrieved. Signed-off-by: Gregory CLEMENT <gregory.clement(a)bootlin.com> Signed-off-by: Thomas Bogendoerfer <tsbogend(a)alpha.franken.de> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- arch/mips/include/asm/mips-cm.h | 22 ++++++++++++++++++++++ arch/mips/kernel/mips-cm.c | 14 ++++++++++++++ 2 files changed, 36 insertions(+) diff --git a/arch/mips/include/asm/mips-cm.h b/arch/mips/include/asm/mips-cm.h index 696b40beb774f..0f31324998c0a 100644 --- a/arch/mips/include/asm/mips-cm.h +++ b/arch/mips/include/asm/mips-cm.h @@ -47,6 +47,16 @@ extern phys_addr_t __mips_cm_phys_base(void); */ extern int mips_cm_is64; +/* + * mips_cm_is_l2_hci_broken - determine if HCI is broken + * + * Some CM reports show that Hardware Cache Initialization is + * complete, but in reality it's not the case. They also incorrectly + * indicate that Hardware Cache Initialization is supported. This + * flags allows warning about this broken feature. + */ +extern bool mips_cm_is_l2_hci_broken; + /** * mips_cm_error_report - Report CM cache errors */ @@ -85,6 +95,18 @@ static inline bool mips_cm_present(void) #endif } +/** + * mips_cm_update_property - update property from the device tree + * + * Retrieve the properties from the device tree if a CM node exist and + * update the internal variable based on this. + */ +#ifdef CONFIG_MIPS_CM +extern void mips_cm_update_property(void); +#else +static void mips_cm_update_property(void) {} +#endif + /** * mips_cm_has_l2sync - determine whether an L2-only sync region is present * diff --git a/arch/mips/kernel/mips-cm.c b/arch/mips/kernel/mips-cm.c index 72c8374a39002..a0d9cde26dc5b 100644 --- a/arch/mips/kernel/mips-cm.c +++ b/arch/mips/kernel/mips-cm.c @@ -5,6 +5,7 @@ */ #include <linux/errno.h> +#include <linux/of.h> #include <linux/percpu.h> #include <linux/spinlock.h> @@ -14,6 +15,7 @@ void __iomem *mips_gcr_base; void __iomem *mips_cm_l2sync_base; int mips_cm_is64; +bool mips_cm_is_l2_hci_broken; static char *cm2_tr[8] = { "mem", "gcr", "gic", "mmio", @@ -238,6 +240,18 @@ static void mips_cm_probe_l2sync(void) mips_cm_l2sync_base = ioremap(addr, MIPS_CM_L2SYNC_SIZE); } +void mips_cm_update_property(void) +{ + struct device_node *cm_node; + + cm_node = of_find_compatible_node(of_root, NULL, "mobileye,eyeq6-cm"); + if (!cm_node) + return; + pr_info("HCI (Hardware Cache Init for the L2 cache) in GCR_L2_RAM_CONFIG from the CM3 is broken"); + mips_cm_is_l2_hci_broken = true; + of_node_put(cm_node); +} + int mips_cm_probe(void) { phys_addr_t addr; -- 2.39.5

3 months

1
5
0 0

[PATCH AUTOSEL 5.15 1/8] MIPS: cm: Detect CM quirks from device tree

by Sasha Levin

From: Gregory CLEMENT <gregory.clement(a)bootlin.com> [ Upstream commit e27fbe16af5cfc40639de4ced67d1a866a1953e9 ] Some information that should be retrieved at runtime for the Coherence Manager can be either absent or wrong. This patch allows checking if some of this information is available from the device tree and updates the internal variable accordingly. For now, only the compatible string associated with the broken HCI is being retrieved. Signed-off-by: Gregory CLEMENT <gregory.clement(a)bootlin.com> Signed-off-by: Thomas Bogendoerfer <tsbogend(a)alpha.franken.de> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- arch/mips/include/asm/mips-cm.h | 22 ++++++++++++++++++++++ arch/mips/kernel/mips-cm.c | 14 ++++++++++++++ 2 files changed, 36 insertions(+) diff --git a/arch/mips/include/asm/mips-cm.h b/arch/mips/include/asm/mips-cm.h index 696b40beb774f..0f31324998c0a 100644 --- a/arch/mips/include/asm/mips-cm.h +++ b/arch/mips/include/asm/mips-cm.h @@ -47,6 +47,16 @@ extern phys_addr_t __mips_cm_phys_base(void); */ extern int mips_cm_is64; +/* + * mips_cm_is_l2_hci_broken - determine if HCI is broken + * + * Some CM reports show that Hardware Cache Initialization is + * complete, but in reality it's not the case. They also incorrectly + * indicate that Hardware Cache Initialization is supported. This + * flags allows warning about this broken feature. + */ +extern bool mips_cm_is_l2_hci_broken; + /** * mips_cm_error_report - Report CM cache errors */ @@ -85,6 +95,18 @@ static inline bool mips_cm_present(void) #endif } +/** + * mips_cm_update_property - update property from the device tree + * + * Retrieve the properties from the device tree if a CM node exist and + * update the internal variable based on this. + */ +#ifdef CONFIG_MIPS_CM +extern void mips_cm_update_property(void); +#else +static void mips_cm_update_property(void) {} +#endif + /** * mips_cm_has_l2sync - determine whether an L2-only sync region is present * diff --git a/arch/mips/kernel/mips-cm.c b/arch/mips/kernel/mips-cm.c index b4f7d950c8468..e21c2fd761674 100644 --- a/arch/mips/kernel/mips-cm.c +++ b/arch/mips/kernel/mips-cm.c @@ -5,6 +5,7 @@ */ #include <linux/errno.h> +#include <linux/of.h> #include <linux/percpu.h> #include <linux/spinlock.h> @@ -14,6 +15,7 @@ void __iomem *mips_gcr_base; void __iomem *mips_cm_l2sync_base; int mips_cm_is64; +bool mips_cm_is_l2_hci_broken; static char *cm2_tr[8] = { "mem", "gcr", "gic", "mmio", @@ -238,6 +240,18 @@ static void mips_cm_probe_l2sync(void) mips_cm_l2sync_base = ioremap(addr, MIPS_CM_L2SYNC_SIZE); } +void mips_cm_update_property(void) +{ + struct device_node *cm_node; + + cm_node = of_find_compatible_node(of_root, NULL, "mobileye,eyeq6-cm"); + if (!cm_node) + return; + pr_info("HCI (Hardware Cache Init for the L2 cache) in GCR_L2_RAM_CONFIG from the CM3 is broken"); + mips_cm_is_l2_hci_broken = true; + of_node_put(cm_node); +} + int mips_cm_probe(void) { phys_addr_t addr; -- 2.39.5

3 months

1
7
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror April 2025