April 2025 - Linux-stable-mirror

[PATCH 5.15.y] scsi: lpfc: Fix null pointer dereference after failing to issue FLOGI and PLOGI

by bin.lan.cn＠windriver.com

From: James Smart <jsmart2021(a)gmail.com> [ Upstream commit 577a942df3de2666f6947bdd3a5c9e8d30073424 ] If lpfc_issue_els_flogi() fails and returns non-zero status, the node reference count is decremented to trigger the release of the nodelist structure. However, if there is a prior registration or dev-loss-evt work pending, the node may be released prematurely. When dev-loss-evt completes, the released node is referenced causing a use-after-free null pointer dereference. Similarly, when processing non-zero ELS PLOGI completion status in lpfc_cmpl_els_plogi(), the ndlp flags are checked for a transport registration before triggering node removal. If dev-loss-evt work is pending, the node may be released prematurely and a subsequent call to lpfc_dev_loss_tmo_handler() results in a use after free ndlp dereference. Add test for pending dev-loss before decrementing the node reference count for FLOGI, PLOGI, PRLI, and ADISC handling. Link: https://lore.kernel.org/r/20220412222008.126521-9-jsmart2021@gmail.com Co-developed-by: Justin Tee <justin.tee(a)broadcom.com> Signed-off-by: Justin Tee <justin.tee(a)broadcom.com> Signed-off-by: James Smart <jsmart2021(a)gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> [Minor context change fixed.] Signed-off-by: Bin Lan <bin.lan.cn(a)windriver.com> Signed-off-by: He Zhe <zhe.he(a)windriver.com> --- Build test passed. --- drivers/scsi/lpfc/lpfc_els.c | 51 +++++++++++++++++++++++++----------- 1 file changed, 35 insertions(+), 16 deletions(-) diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c index 5f44a0763f37..134d56bd00da 100644 --- a/drivers/scsi/lpfc/lpfc_els.c +++ b/drivers/scsi/lpfc/lpfc_els.c @@ -1517,10 +1517,13 @@ lpfc_initial_flogi(struct lpfc_vport *vport) } if (lpfc_issue_els_flogi(vport, ndlp, 0)) { - /* This decrement of reference count to node shall kick off - * the release of the node. + /* A node reference should be retained while registered with a + * transport or dev-loss-evt work is pending. + * Otherwise, decrement node reference to trigger release. */ - lpfc_nlp_put(ndlp); + if (!(ndlp->fc4_xpt_flags & (SCSI_XPT_REGD | NVME_XPT_REGD)) && + !(ndlp->nlp_flag & NLP_IN_DEV_LOSS)) + lpfc_nlp_put(ndlp); return 0; } return 1; @@ -1563,10 +1566,13 @@ lpfc_initial_fdisc(struct lpfc_vport *vport) } if (lpfc_issue_els_fdisc(vport, ndlp, 0)) { - /* decrement node reference count to trigger the release of - * the node. + /* A node reference should be retained while registered with a + * transport or dev-loss-evt work is pending. + * Otherwise, decrement node reference to trigger release. */ - lpfc_nlp_put(ndlp); + if (!(ndlp->fc4_xpt_flags & (SCSI_XPT_REGD | NVME_XPT_REGD)) && + !(ndlp->nlp_flag & NLP_IN_DEV_LOSS)) + lpfc_nlp_put(ndlp); return 0; } return 1; @@ -1967,6 +1973,7 @@ lpfc_cmpl_els_plogi(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, struct lpfc_dmabuf *prsp; int disc; struct serv_parm *sp = NULL; + bool release_node = false; /* we pass cmdiocb to state machine which needs rspiocb as well */ cmdiocb->context_un.rsp_iocb = rspiocb; @@ -2047,19 +2054,21 @@ lpfc_cmpl_els_plogi(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, spin_unlock_irq(&ndlp->lock); goto out; } - spin_unlock_irq(&ndlp->lock); /* No PLOGI collision and the node is not registered with the * scsi or nvme transport. It is no longer an active node. Just * start the device remove process. */ if (!(ndlp->fc4_xpt_flags & (SCSI_XPT_REGD | NVME_XPT_REGD))) { - spin_lock_irq(&ndlp->lock); ndlp->nlp_flag &= ~NLP_NPR_2B_DISC; - spin_unlock_irq(&ndlp->lock); + if (!(ndlp->nlp_flag & NLP_IN_DEV_LOSS)) + release_node = true; + } + spin_unlock_irq(&ndlp->lock); + + if (release_node) lpfc_disc_state_machine(vport, ndlp, cmdiocb, NLP_EVT_DEVICE_RM); - } } else { /* Good status, call state machine */ prsp = list_entry(((struct lpfc_dmabuf *) @@ -2269,6 +2278,7 @@ lpfc_cmpl_els_prli(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, struct lpfc_nodelist *ndlp; char *mode; u32 loglevel; + bool release_node = false; /* we pass cmdiocb to state machine which needs rspiocb as well */ cmdiocb->context_un.rsp_iocb = rspiocb; @@ -2335,14 +2345,18 @@ lpfc_cmpl_els_prli(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, * it is no longer an active node. Otherwise devloss * handles the final cleanup. */ + spin_lock_irq(&ndlp->lock); if (!(ndlp->fc4_xpt_flags & (SCSI_XPT_REGD | NVME_XPT_REGD)) && !ndlp->fc4_prli_sent) { - spin_lock_irq(&ndlp->lock); ndlp->nlp_flag &= ~NLP_NPR_2B_DISC; - spin_unlock_irq(&ndlp->lock); + if (!(ndlp->nlp_flag & NLP_IN_DEV_LOSS)) + release_node = true; + } + spin_unlock_irq(&ndlp->lock); + + if (release_node) lpfc_disc_state_machine(vport, ndlp, cmdiocb, NLP_EVT_DEVICE_RM); - } } else { /* Good status, call state machine. However, if another * PRLI is outstanding, don't call the state machine @@ -2713,6 +2727,7 @@ lpfc_cmpl_els_adisc(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, IOCB_t *irsp; struct lpfc_nodelist *ndlp; int disc; + bool release_node = false; /* we pass cmdiocb to state machine which needs rspiocb as well */ cmdiocb->context_un.rsp_iocb = rspiocb; @@ -2771,13 +2786,17 @@ lpfc_cmpl_els_adisc(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, * transport, it is no longer an active node. Otherwise * devloss handles the final cleanup. */ + spin_lock_irq(&ndlp->lock); if (!(ndlp->fc4_xpt_flags & (SCSI_XPT_REGD | NVME_XPT_REGD))) { - spin_lock_irq(&ndlp->lock); ndlp->nlp_flag &= ~NLP_NPR_2B_DISC; - spin_unlock_irq(&ndlp->lock); + if (!(ndlp->nlp_flag & NLP_IN_DEV_LOSS)) + release_node = true; + } + spin_unlock_irq(&ndlp->lock); + + if (release_node) lpfc_disc_state_machine(vport, ndlp, cmdiocb, NLP_EVT_DEVICE_RM); - } } else /* Good status, call state machine */ lpfc_disc_state_machine(vport, ndlp, cmdiocb, -- 2.34.1

1 week, 1 day

2
1
0 0

[REGRESSION] stable-rc/linux-5.15.y: (build) variable 'is_redirect' is used uninitialized whenever 'if' conditi...

by KernelCI bot

Hello, New build issue found on stable-rc/linux-5.15.y: --- variable 'is_redirect' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized] in net/sched/act_mirred.o (net/sched/act_mirred.c) [logspec:kbuild,kbuild.compiler.error] --- - dashboard: https://d.kernelci.org/i/maestro:72f9cc14e764580fd20bda28956c2ddf82023571 - giturl: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git - commit HEAD: dd688b41d9038dbc86718bae12ed995d596c1de5 Log excerpt: ===================================================== net/sched/act_mirred.c:264:6: error: variable 'is_redirect' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized] 264 | if (unlikely(!(dev->flags & IFF_UP)) || !netif_carrier_ok(dev)) { | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ./include/linux/compiler.h:78:22: note: expanded from macro 'unlikely' 78 | # define unlikely(x) __builtin_expect(!!(x), 0) | ^ net/sched/act_mirred.c:331:6: note: uninitialized use occurs here 331 | if (is_redirect) | ^~~~~~~~~~~ net/sched/act_mirred.c:264:2: note: remove the 'if' if its condition is always false 264 | if (unlikely(!(dev->flags & IFF_UP)) || !netif_carrier_ok(dev)) { | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 265 | net_notice_ratelimited("tc mirred to Houston: device %s is down\n", | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 266 | dev->name); | ~~~~~~~~~~~ 267 | goto err_cant_do; | ~~~~~~~~~~~~~~~~~ 268 | } | ~ net/sched/act_mirred.c:264:6: error: variable 'is_redirect' is used uninitialized whenever '||' condition is true [-Werror,-Wsometimes-uninitialized] 264 | if (unlikely(!(dev->flags & IFF_UP)) || !netif_carrier_ok(dev)) { | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ./include/linux/compiler.h:78:22: note: expanded from macro 'unlikely' 78 | # define unlikely(x) __builtin_expect(!!(x), 0) | ^~~~~~~~~~~~~~~~~~~~~~~~~~ net/sched/act_mirred.c:331:6: note: uninitialized use occurs here 331 | if (is_redirect) | ^~~~~~~~~~~ net/sched/act_mirred.c:264:6: note: remove the '||' if its condition is always false 264 | if (unlikely(!(dev->flags & IFF_UP)) || !netif_carrier_ok(dev)) { | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ./include/linux/compiler.h:78:22: note: expanded from macro 'unlikely' 78 | # define unlikely(x) __builtin_expect(!!(x), 0) | ^ net/sched/act_mirred.c:259:6: error: variable 'is_redirect' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized] 259 | if (unlikely(!dev)) { | ^~~~~~~~~~~~~~ ./include/linux/compiler.h:78:22: note: expanded from macro 'unlikely' 78 | # define unlikely(x) __builtin_expect(!!(x), 0) | ^~~~~~~~~~~~~~~~~~~~~~~~~~ net/sched/act_mirred.c:331:6: note: uninitialized use occurs here 331 | if (is_redirect) | ^~~~~~~~~~~ net/sched/act_mirred.c:259:2: note: remove the 'if' if its condition is always false 259 | if (unlikely(!dev)) { | ^~~~~~~~~~~~~~~~~~~~~ 260 | pr_notice_once("tc mirred: target device is gone\n"); | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 261 | goto err_cant_do; | ~~~~~~~~~~~~~~~~~ 262 | } | ~ net/sched/act_mirred.c:237:18: note: initialize the variable 'is_redirect' to silence this warning 237 | bool is_redirect; | ^ | = 0 3 errors generated. ===================================================== # Builds where the incident occurred: ## defconfig+allmodconfig+CONFIG_FRAME_WARN=2048 on (arm): - compiler: clang-17 - dashboard: https://d.kernelci.org/build/maestro:6808fe8b43948caad9509059 #kernelci issue maestro:72f9cc14e764580fd20bda28956c2ddf82023571 Reported-by: kernelci.org bot <bot(a)kernelci.org> -- This is an experimental report format. Please send feedback in! Talk to us at kernelci(a)lists.linux.dev Made with love by the KernelCI team - https://kernelci.org

1 week, 1 day

1
0
0 0

[PATCH v2 1/2] drm/nouveau: Fix WARN_ON in nouveau_fence_context_kill()

by Philipp Stanner

Nouveau is mostly designed in a way that it's expected that fences only ever get signaled through nouveau_fence_signal(). However, in at least one other place, nouveau_fence_done(), can signal fences, too. If that happens (race) a signaled fence remains in the pending list for a while, until it gets removed by nouveau_fence_update(). Should nouveau_fence_context_kill() run in the meantime, this would be a bug because the function would attempt to set an error code on an already signaled fence. Have nouveau_fence_context_kill() check for a fence being signaled. Cc: <stable(a)vger.kernel.org> # v5.10+ Fixes: ea13e5abf807 ("drm/nouveau: signal pending fences when channel has been killed") Suggested-by: Christian König <christian.koenig(a)amd.com> Signed-off-by: Philipp Stanner <phasta(a)kernel.org> --- drivers/gpu/drm/nouveau/nouveau_fence.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c index 7622587f149e..6ded8c2b6d3b 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.c +++ b/drivers/gpu/drm/nouveau/nouveau_fence.c @@ -90,7 +90,7 @@ nouveau_fence_context_kill(struct nouveau_fence_chan *fctx, int error) while (!list_empty(&fctx->pending)) { fence = list_entry(fctx->pending.next, typeof(*fence), head); - if (error) + if (error && !dma_fence_is_signaled_locked(&fence->base)) dma_fence_set_error(&fence->base, error); if (nouveau_fence_signal(fence)) -- 2.48.1

1 week, 1 day

2
1
0 0

Potential Linux Crash: WARNING in ext4_mb_load_buddy_gfp in linux6.12.24(longterm maintenance)

by ffhgfv

Hello, I found a potential bug titled " WARNING in ext4_mb_load_buddy_gfp " with modified syzkaller in the Linux6.12.24(longterm maintenance, last updated on April 20, 2025). It seems to be a problem with the ext4 subsystem. Unfortunately, I am unable to reproduce this bug. If you fix this issue, please add the following tag to the commit: Reported-by: Jianzhou Zhao <xnxc22xnxc22(a)qq.com>, xingwei lee <xrivendell7(a)gmail.com>,Penglei Jiang <superman.xpt(a)gmail.com> The commit of the kernel is : b6efa8ce222e58cfe2bbaa4e3329818c2b4bd74e kernel config: https://syzkaller.appspot.com/text?tag=KernelConfig&x=55f8591b98dd132 compiler: gcc version 11.4.0 ------------[ cut here ]----------------------------------------- TITLE: WARNING in ext4_mb_load_buddy_gfp ------------[ cut here ]------------ WARNING: CPU: 0 PID: 96 at mm/page_alloc.c:4240 __alloc_pages_slowpath mm/page_alloc.c:4240 [inline] WARNING: CPU: 0 PID: 96 at mm/page_alloc.c:4240 __alloc_pages_noprof+0x1a1d/0x21f0 mm/page_alloc.c:4767 Modules linked in: CPU: 0 UID: 0 PID: 96 Comm: kswapd0 Not tainted 6.12.24 #3 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 RIP: 0010:__alloc_pages_slowpath mm/page_alloc.c:4240 [inline] RIP: 0010:__alloc_pages_noprof+0x1a1d/0x21f0 mm/page_alloc.c:4767 Code: 89 fa 48 c1 ea 03 0f b6 14 02 48 89 f8 83 e0 07 83 c0 03 38 d0 7c 08 84 d2 0f 85 a2 07 00 00 f6 43 2d 08 0f 84 bd ea ff ff 90 <0f> 0b 90 e9 b4 ea ff ff 65 8b 15 70 e5 23 7e 83 fa 07 0f 87 56 06 RSP: 0000:ffffc90001376e18 EFLAGS: 00010202 RAX: 0000000000000007 RBX: ffff888042ac4a00 RCX: ffffc90001376f44 RDX: 0000000000000000 RSI: 1ffff9200026edb0 RDI: ffff888042ac4a2c RBP: ffff88803fffbc80 R08: 000000000000699a R09: 0000000000000400 R10: ffff88807ffda357 R11: 0000000000000000 R12: 0000000000000000 R13: ffffffffffffffff R14: 0000000000048c40 R15: 0000000000000001 FS: 0000000000000000(0000) GS:ffff88802b800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f4969a4e7b0 CR3: 000000002719c000 CR4: 0000000000752ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: <task> alloc_pages_mpol_noprof+0x2b6/0x600 mm/mempolicy.c:2269 folio_alloc_noprof+0x21/0xc0 mm/mempolicy.c:2356 filemap_alloc_folio_noprof+0x3d2/0x470 mm/filemap.c:1010 __filemap_get_folio+0x57f/0xb00 mm/filemap.c:1952 ext4_mb_load_buddy_gfp+0x74d/0x10c0 fs/ext4/mballoc.c:1640 ext4_discard_preallocations+0x539/0xf00 fs/ext4/mballoc.c:5592 ext4_clear_inode+0x3d/0x230 fs/ext4/super.c:1523 ext4_evict_inode+0x243/0x18c0 fs/ext4/inode.c:323 evict+0x3ef/0x940 fs/inode.c:725 dispose_list+0x117/0x1e0 fs/inode.c:774 prune_icache_sb+0xeb/0x150 fs/inode.c:963 super_cache_scan+0x37f/0x570 fs/super.c:223 do_shrink_slab+0x44b/0x1190 mm/shrinker.c:437 shrink_slab_memcg mm/shrinker.c:550 [inline] shrink_slab+0xb61/0x12a0 mm/shrinker.c:628 shrink_one+0x4ad/0x7c0 mm/vmscan.c:4835 shrink_many mm/vmscan.c:4896 [inline] lru_gen_shrink_node mm/vmscan.c:4974 [inline] shrink_node+0x2420/0x3890 mm/vmscan.c:5954 kswapd_shrink_node mm/vmscan.c:6782 [inline] balance_pgdat+0xbe5/0x18c0 mm/vmscan.c:6974 kswapd+0x702/0xd50 mm/vmscan.c:7243 kthread+0x2c7/0x3b0 kernel/kthread.c:389 ret_from_fork+0x45/0x80 arch/x86/kernel/process.c:152 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244 </task> ================================================================== I hope it helps. Best regards Jianzhou Zhao</superman.xpt(a)gmail.com></xrivendell7(a)gmail.com></xnxc22xnxc22(a)qq.com>

1 week, 1 day

1
0
0 0

[tip: x86/urgent] x86/mm: Fix _pgd_alloc() for Xen PV mode

by tip-bot2 for Juergen Gross

The following commit has been merged into the x86/urgent branch of tip: Commit-ID: 4ce385f56434f3810ef103e1baea357ddcc6667e Gitweb: https://git.kernel.org/tip/4ce385f56434f3810ef103e1baea357ddcc6667e Author: Juergen Gross <jgross(a)suse.com> AuthorDate: Tue, 22 Apr 2025 15:17:17 +02:00 Committer: Dave Hansen <dave.hansen(a)linux.intel.com> CommitterDate: Wed, 23 Apr 2025 07:49:14 -07:00 x86/mm: Fix _pgd_alloc() for Xen PV mode Recently _pgd_alloc() was switched from using __get_free_pages() to pagetable_alloc_noprof(), which might return a compound page in case the allocation order is larger than 0. On x86 this will be the case if CONFIG_MITIGATION_PAGE_TABLE_ISOLATION is set, even if PTI has been disabled at runtime. When running as a Xen PV guest (this will always disable PTI), using a compound page for a PGD will result in VM_BUG_ON_PGFLAGS being triggered when the Xen code tries to pin the PGD. Fix the Xen issue together with the not needed 8k allocation for a PGD with PTI disabled by replacing PGD_ALLOCATION_ORDER with an inline helper returning the needed order for PGD allocations. Fixes: a9b3c355c2e6 ("asm-generic: pgalloc: provide generic __pgd_{alloc,free}") Reported-by: Petr Vaněk <arkamar(a)atlas.cz> Signed-off-by: Juergen Gross <jgross(a)suse.com> Signed-off-by: Dave Hansen <dave.hansen(a)linux.intel.com> Tested-by: Petr Vaněk <arkamar(a)atlas.cz> Cc:stable@vger.kernel.org Link: https://lore.kernel.org/all/20250422131717.25724-1-jgross%40suse.com --- arch/x86/include/asm/pgalloc.h | 19 +++++++++++-------- arch/x86/kernel/machine_kexec_32.c | 4 ++-- arch/x86/mm/pgtable.c | 4 ++-- arch/x86/platform/efi/efi_64.c | 4 ++-- 4 files changed, 17 insertions(+), 14 deletions(-) diff --git a/arch/x86/include/asm/pgalloc.h b/arch/x86/include/asm/pgalloc.h index a331475..c88691b 100644 --- a/arch/x86/include/asm/pgalloc.h +++ b/arch/x86/include/asm/pgalloc.h @@ -6,6 +6,8 @@ #include <linux/mm.h> /* for struct page */ #include <linux/pagemap.h> +#include <asm/cpufeature.h> + #define __HAVE_ARCH_PTE_ALLOC_ONE #define __HAVE_ARCH_PGD_FREE #include <asm-generic/pgalloc.h> @@ -29,16 +31,17 @@ static inline void paravirt_release_pud(unsigned long pfn) {} static inline void paravirt_release_p4d(unsigned long pfn) {} #endif -#ifdef CONFIG_MITIGATION_PAGE_TABLE_ISOLATION /* - * Instead of one PGD, we acquire two PGDs. Being order-1, it is - * both 8k in size and 8k-aligned. That lets us just flip bit 12 - * in a pointer to swap between the two 4k halves. + * In case of Page Table Isolation active, we acquire two PGDs instead of one. + * Being order-1, it is both 8k in size and 8k-aligned. That lets us just + * flip bit 12 in a pointer to swap between the two 4k halves. */ -#define PGD_ALLOCATION_ORDER 1 -#else -#define PGD_ALLOCATION_ORDER 0 -#endif +static inline unsigned int pgd_allocation_order(void) +{ + if (cpu_feature_enabled(X86_FEATURE_PTI)) + return 1; + return 0; +} /* * Allocate and free page tables. diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c index 8026516..1f32530 100644 --- a/arch/x86/kernel/machine_kexec_32.c +++ b/arch/x86/kernel/machine_kexec_32.c @@ -42,7 +42,7 @@ static void load_segments(void) static void machine_kexec_free_page_tables(struct kimage *image) { - free_pages((unsigned long)image->arch.pgd, PGD_ALLOCATION_ORDER); + free_pages((unsigned long)image->arch.pgd, pgd_allocation_order()); image->arch.pgd = NULL; #ifdef CONFIG_X86_PAE free_page((unsigned long)image->arch.pmd0); @@ -59,7 +59,7 @@ static void machine_kexec_free_page_tables(struct kimage *image) static int machine_kexec_alloc_page_tables(struct kimage *image) { image->arch.pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, - PGD_ALLOCATION_ORDER); + pgd_allocation_order()); #ifdef CONFIG_X86_PAE image->arch.pmd0 = (pmd_t *)get_zeroed_page(GFP_KERNEL); image->arch.pmd1 = (pmd_t *)get_zeroed_page(GFP_KERNEL); diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c index a05fcdd..f7ae44d 100644 --- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -360,7 +360,7 @@ static inline pgd_t *_pgd_alloc(struct mm_struct *mm) * We allocate one page for pgd. */ if (!SHARED_KERNEL_PMD) - return __pgd_alloc(mm, PGD_ALLOCATION_ORDER); + return __pgd_alloc(mm, pgd_allocation_order()); /* * Now PAE kernel is not running as a Xen domain. We can allocate @@ -380,7 +380,7 @@ static inline void _pgd_free(struct mm_struct *mm, pgd_t *pgd) static inline pgd_t *_pgd_alloc(struct mm_struct *mm) { - return __pgd_alloc(mm, PGD_ALLOCATION_ORDER); + return __pgd_alloc(mm, pgd_allocation_order()); } static inline void _pgd_free(struct mm_struct *mm, pgd_t *pgd) diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index ac57259..a4b4ebd 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -73,7 +73,7 @@ int __init efi_alloc_page_tables(void) gfp_t gfp_mask; gfp_mask = GFP_KERNEL | __GFP_ZERO; - efi_pgd = (pgd_t *)__get_free_pages(gfp_mask, PGD_ALLOCATION_ORDER); + efi_pgd = (pgd_t *)__get_free_pages(gfp_mask, pgd_allocation_order()); if (!efi_pgd) goto fail; @@ -96,7 +96,7 @@ free_p4d: if (pgtable_l5_enabled()) free_page((unsigned long)pgd_page_vaddr(*pgd)); free_pgd: - free_pages((unsigned long)efi_pgd, PGD_ALLOCATION_ORDER); + free_pages((unsigned long)efi_pgd, pgd_allocation_order()); fail: return -ENOMEM; }

1 week, 1 day

1
0
0 0

[PATCH] fs/nfs/read: fix double-unlock bug in nfs_return_empty_folio()

by Max Kellermann

Sometimes, when a file was read while it was being truncated by another NFS client, the kernel could deadlock because folio_unlock() was called twice, and the second call would XOR back the `PG_locked` flag. Most of the time (depending on the timing of the truncation), nobody notices the problem because folio_unlock() gets called three times, which flips `PG_locked` back off: 1. vfs_read, nfs_read_folio, ... nfs_read_add_folio, nfs_return_empty_folio 2. vfs_read, nfs_read_folio, ... netfs_read_collection, netfs_unlock_abandoned_read_pages 3. vfs_read, ... nfs_do_read_folio, nfs_read_add_folio, nfs_return_empty_folio The problem is that nfs_read_add_folio() is not supposed to unlock the folio if fscache is enabled, and a nfs_netfs_folio_unlock() check is missing in nfs_return_empty_folio(). Rarely this leads to a warning in netfs_read_collection(): ------------[ cut here ]------------ R=0000031c: folio 10 is not locked WARNING: CPU: 0 PID: 29 at fs/netfs/read_collect.c:133 netfs_read_collection+0x7c0/0xf00 [...] Workqueue: events_unbound netfs_read_collection_worker RIP: 0010:netfs_read_collection+0x7c0/0xf00 [...] Call Trace: <TASK> netfs_read_collection_worker+0x67/0x80 process_one_work+0x12e/0x2c0 worker_thread+0x295/0x3a0 Most of the time, however, processes just get stuck forever in folio_wait_bit_common(), waiting for `PG_locked` to disappear, which never happens because nobody is really holding the folio lock. Fixes: 000dbe0bec05 ("NFS: Convert buffered read paths to use netfs when fscache is enabled") Cc: stable(a)vger.kernel.org Signed-off-by: Max Kellermann <max.kellermann(a)ionos.com> --- fs/nfs/read.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 81bd1b9aba17..3c1fa320b3f1 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -56,7 +56,8 @@ static int nfs_return_empty_folio(struct folio *folio) { folio_zero_segment(folio, 0, folio_size(folio)); folio_mark_uptodate(folio); - folio_unlock(folio); + if (nfs_netfs_folio_unlock(folio)) + folio_unlock(folio); return 0; } -- 2.47.2

1 week, 1 day

2
1
0 0

[PATCH v5.4.y] usb: dwc3: core: Do core softreset when switch mode

by Hardik Gohil

From: Yu Chen <chenyu56(a)huawei.com> commit f88359e1588b85cf0e8209ab7d6620085f3441d9 upstream. From: John Stultz <john.stultz(a)linaro.org> According to the programming guide, to switch mode for DRD controller, the driver needs to do the following. To switch from device to host: 1. Reset controller with GCTL.CoreSoftReset 2. Set GCTL.PrtCapDir(host mode) 3. Reset the host with USBCMD.HCRESET 4. Then follow up with the initializing host registers sequence To switch from host to device: 1. Reset controller with GCTL.CoreSoftReset 2. Set GCTL.PrtCapDir(device mode) 3. Reset the device with DCTL.CSftRst 4. Then follow up with the initializing registers sequence Currently we're missing step 1) to do GCTL.CoreSoftReset and step 3) of switching from host to device. John Stult reported a lockup issue seen with HiKey960 platform without these steps[1]. Similar issue is observed with Ferry's testing platform[2]. So, apply the required steps along with some fixes to Yu Chen's and John Stultz's version. The main fixes to their versions are the missing wait for clocks synchronization before clearing GCTL.CoreSoftReset and only apply DCTL.CSftRst when switching from host to device. [1] https://lore.kernel.org/linux-usb/20210108015115.27920-1-john.stultz@linaro… [2] https://lore.kernel.org/linux-usb/0ba7a6ba-e6a7-9cd4-0695-64fc927e01f1@gmai… Fixes: 41ce1456e1db ("usb: dwc3: core: make dwc3_set_mode() work properly") Cc: Andy Shevchenko <andy.shevchenko(a)gmail.com> Cc: Ferry Toth <fntoth(a)gmail.com> Cc: Wesley Cheng <wcheng(a)codeaurora.org> Cc: <stable(a)vger.kernel.org> Tested-by: John Stultz <john.stultz(a)linaro.org> Tested-by: Wesley Cheng <wcheng(a)codeaurora.org> Signed-off-by: Yu Chen <chenyu56(a)huawei.com> Signed-off-by: John Stultz <john.stultz(a)linaro.org> Signed-off-by: Thinh Nguyen <Thinh.Nguyen(a)synopsys.com> Link: https://lore.kernel.org/r/374440f8dcd4f06c02c2caf4b1efde86774e02d9.16185216… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Hardik Gohil <hgohil(a)mvista.com> --- this fix is missing in v5.4.y stable version apply the following dependend patch before applying this patch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=… drivers/usb/dwc3/core.c | 25 +++++++++++++++++++++++++ drivers/usb/dwc3/core.h | 5 +++++ 2 files changed, 30 insertions(+) diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c index 1a420c00d6ca..650eb4f735f9 100644 --- a/drivers/usb/dwc3/core.c +++ b/drivers/usb/dwc3/core.c @@ -122,6 +122,8 @@ static void __dwc3_set_mode(struct work_struct *work) if (dwc->dr_mode != USB_DR_MODE_OTG) return; + mutex_lock(&dwc->mutex); + pm_runtime_get_sync(dwc->dev); if (dwc->current_dr_role == DWC3_GCTL_PRTCAP_OTG) @@ -155,6 +157,25 @@ static void __dwc3_set_mode(struct work_struct *work) break; } + /* For DRD host or device mode only */ + if (dwc->desired_dr_role != DWC3_GCTL_PRTCAP_OTG) { + reg = dwc3_readl(dwc->regs, DWC3_GCTL); + reg |= DWC3_GCTL_CORESOFTRESET; + dwc3_writel(dwc->regs, DWC3_GCTL, reg); + + /* + * Wait for internal clocks to synchronized. DWC_usb31 and + * DWC_usb32 may need at least 50ms (less for DWC_usb3). To + * keep it consistent across different IPs, let's wait up to + * 100ms before clearing GCTL.CORESOFTRESET. + */ + msleep(100); + + reg = dwc3_readl(dwc->regs, DWC3_GCTL); + reg &= ~DWC3_GCTL_CORESOFTRESET; + dwc3_writel(dwc->regs, DWC3_GCTL, reg); + } + spin_lock_irqsave(&dwc->lock, flags); dwc3_set_prtcap(dwc, dwc->desired_dr_role); @@ -179,6 +200,8 @@ static void __dwc3_set_mode(struct work_struct *work) } break; case DWC3_GCTL_PRTCAP_DEVICE: + dwc3_core_soft_reset(dwc); + dwc3_event_buffers_setup(dwc); if (dwc->usb2_phy) @@ -201,6 +224,7 @@ static void __dwc3_set_mode(struct work_struct *work) out: pm_runtime_mark_last_busy(dwc->dev); pm_runtime_put_autosuspend(dwc->dev); + mutex_unlock(&dwc->mutex); } void dwc3_set_mode(struct dwc3 *dwc, u32 mode) @@ -1511,6 +1535,7 @@ static int dwc3_probe(struct platform_device *pdev) dwc3_cache_hwparams(dwc); spin_lock_init(&dwc->lock); + mutex_init(&dwc->mutex); pm_runtime_get_noresume(dev); pm_runtime_set_active(dev); diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h index 34f3fbba391b..44b0239676a1 100644 --- a/drivers/usb/dwc3/core.h +++ b/drivers/usb/dwc3/core.h @@ -13,6 +13,7 @@ #include <linux/device.h> #include <linux/spinlock.h> +#include <linux/mutex.h> #include <linux/ioport.h> #include <linux/list.h> #include <linux/bitops.h> @@ -929,6 +930,7 @@ struct dwc3_scratchpad_array { * @scratch_addr: dma address of scratchbuf * @ep0_in_setup: one control transfer is completed and enter setup phase * @lock: for synchronizing + * @mutex: for mode switching * @dev: pointer to our struct device * @sysdev: pointer to the DMA-capable device * @xhci: pointer to our xHCI child @@ -1061,6 +1063,9 @@ struct dwc3 { /* device lock */ spinlock_t lock; + /* mode switching lock */ + struct mutex mutex; + struct device *dev; struct device *sysdev; -- 2.25.1

1 week, 1 day

3
2
0 0

FAILED: patch "[PATCH] mmc: sdhci-brcmstb: add cqhci suspend/resume to PM ops" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y git checkout FETCH_HEAD git cherry-pick -x 723ef0e20dbb2aa1b5406d2bb75374fc48187daa # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025032414-unsheathe-greedily-1d17@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 723ef0e20dbb2aa1b5406d2bb75374fc48187daa Mon Sep 17 00:00:00 2001 From: Kamal Dasu <kamal.dasu(a)broadcom.com> Date: Tue, 11 Mar 2025 12:59:35 -0400 Subject: [PATCH] mmc: sdhci-brcmstb: add cqhci suspend/resume to PM ops cqhci timeouts observed on brcmstb platforms during suspend: ... [ 164.832853] mmc0: cqhci: timeout for tag 18 ... Adding cqhci_suspend()/resume() calls to disable cqe in sdhci_brcmstb_suspend()/resume() respectively to fix CQE timeouts seen on PM suspend. Fixes: d46ba2d17f90 ("mmc: sdhci-brcmstb: Add support for Command Queuing (CQE)") Cc: stable(a)vger.kernel.org Signed-off-by: Kamal Dasu <kamal.dasu(a)broadcom.com> Reviewed-by: Florian Fainelli <florian.fainelli(a)broadcom.com> Link: https://lore.kernel.org/r/20250311165946.28190-1-kamal.dasu@broadcom.com Signed-off-by: Ulf Hansson <ulf.hansson(a)linaro.org> diff --git a/drivers/mmc/host/sdhci-brcmstb.c b/drivers/mmc/host/sdhci-brcmstb.c index 0ef4d578ade8..48cdcba0f39c 100644 --- a/drivers/mmc/host/sdhci-brcmstb.c +++ b/drivers/mmc/host/sdhci-brcmstb.c @@ -503,8 +503,15 @@ static int sdhci_brcmstb_suspend(struct device *dev) struct sdhci_host *host = dev_get_drvdata(dev); struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host); struct sdhci_brcmstb_priv *priv = sdhci_pltfm_priv(pltfm_host); + int ret; clk_disable_unprepare(priv->base_clk); + if (host->mmc->caps2 & MMC_CAP2_CQE) { + ret = cqhci_suspend(host->mmc); + if (ret) + return ret; + } + return sdhci_pltfm_suspend(dev); } @@ -529,6 +536,9 @@ static int sdhci_brcmstb_resume(struct device *dev) ret = clk_set_rate(priv->base_clk, priv->base_freq_hz); } + if (host->mmc->caps2 & MMC_CAP2_CQE) + ret = cqhci_resume(host->mmc); + return ret; } #endif

1 week, 1 day

3
7
0 0

[PATCH 0/2] regulator: max20086: Fixes chip id and enable gpio

by João Paulo Gonçalves

Hello, I'm working on integrating a system with a MAX20086 and noticed these small issues in the driver: the chip ID for MAX20086 is 0x30 and not 0x40. Also, in my use case, the enable pin is always enabled by hardware, so the enable GPIO isn't needed. Without these changes, the driver fails to probe. Signed-off-by: João Paulo Gonçalves <jpaulo.silvagoncalves(a)gmail.com> --- João Paulo Gonçalves (2): regulator: max20086: Fix MAX200086 chip id regulator: max20086: Change enable gpio to optional drivers/regulator/max20086-regulator.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) --- base-commit: 6fea5fabd3323cd27b2ab5143263f37ff29550cb change-id: 20250420-fix-max20086-430a63f7a9eb Best regards, -- João Paulo Gonçalves <jpaulo.silvagoncalves(a)gmail.com>

1 week, 1 day

2
3
0 0

[PATCH] usb: typec: ucsi: fix Clang -Wsign-conversion warning

by Qasim Ijaz

debugfs.c emits the following warnings when compiling with the -Wsign-conversion flag with clang 15: drivers/usb/typec/ucsi/debugfs.c:58:27: warning: implicit conversion changes signedness: 'int' to 'u32' (aka 'unsigned int') [-Wsign-conversion] ucsi->debugfs->status = ret; ~ ^~~ drivers/usb/typec/ucsi/debugfs.c:71:25: warning: implicit conversion changes signedness: 'u32' (aka 'unsigned int') to 'int' [-Wsign-conversion] return ucsi->debugfs->status; ~~~~~~ ~~~~~~~~~~~~~~~^~~~~~ During ucsi_cmd() we see: if (ret < 0) { ucsi->debugfs->status = ret; return ret; } But "status" is u32 meaning unsigned wrap-around occurs when assigning a value which is < 0 to it, this obscures the real status. To fix this make the "status" of type int since ret is also of type int. Fixes: df0383ffad64 ("usb: typec: ucsi: Add debugfs for ucsi commands") Cc: stable(a)vger.kernel.org Signed-off-by: Qasim Ijaz <qasdev00(a)gmail.com> --- drivers/usb/typec/ucsi/ucsi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/typec/ucsi/ucsi.h b/drivers/usb/typec/ucsi/ucsi.h index 3a2c1762bec1..525d28160413 100644 --- a/drivers/usb/typec/ucsi/ucsi.h +++ b/drivers/usb/typec/ucsi/ucsi.h @@ -432,7 +432,7 @@ struct ucsi_debugfs_entry { u64 low; u64 high; } response; - u32 status; + int status; struct dentry *dentry; }; -- 2.39.5

1 week, 1 day

2
1
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror April 2025