- Linux-stable-mirror - lists.linaro.org

FAILED: patch "[PATCH] xhci: dbc: Fix full DbC transfer ring after several" failed to apply to 6.12-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.12-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-6.12.y git checkout FETCH_HEAD git cherry-pick -x a5c98e8b1398534ae1feb6e95e2d3ee5215538ed # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025091756-glare-cyclic-9298@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From a5c98e8b1398534ae1feb6e95e2d3ee5215538ed Mon Sep 17 00:00:00 2001 From: Mathias Nyman <mathias.nyman(a)linux.intel.com> Date: Tue, 2 Sep 2025 13:53:05 +0300 Subject: [PATCH] xhci: dbc: Fix full DbC transfer ring after several reconnects Pending requests will be flushed on disconnect, and the corresponding TRBs will be turned into No-op TRBs, which are ignored by the xHC controller once it starts processing the ring. If the USB debug cable repeatedly disconnects before ring is started then the ring will eventually be filled with No-op TRBs. No new transfers can be queued when the ring is full, and driver will print the following error message: "xhci_hcd 0000:00:14.0: failed to queue trbs" This is a normal case for 'in' transfers where TRBs are always enqueued in advance, ready to take on incoming data. If no data arrives, and device is disconnected, then ring dequeue will remain at beginning of the ring while enqueue points to first free TRB after last cancelled No-op TRB. s Solve this by reinitializing the rings when the debug cable disconnects and DbC is leaving the configured state. Clear the whole ring buffer and set enqueue and dequeue to the beginning of ring, and set cycle bit to its initial state. Cc: stable(a)vger.kernel.org Fixes: dfba2174dc42 ("usb: xhci: Add DbC support in xHCI driver") Signed-off-by: Mathias Nyman <mathias.nyman(a)linux.intel.com> Link: https://lore.kernel.org/r/20250902105306.877476-3-mathias.nyman@linux.intel… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> diff --git a/drivers/usb/host/xhci-dbgcap.c b/drivers/usb/host/xhci-dbgcap.c index d0faff233e3e..63edf2d8f245 100644 --- a/drivers/usb/host/xhci-dbgcap.c +++ b/drivers/usb/host/xhci-dbgcap.c @@ -462,6 +462,25 @@ static void xhci_dbc_ring_init(struct xhci_ring *ring) xhci_initialize_ring_info(ring); } +static int xhci_dbc_reinit_ep_rings(struct xhci_dbc *dbc) +{ + struct xhci_ring *in_ring = dbc->eps[BULK_IN].ring; + struct xhci_ring *out_ring = dbc->eps[BULK_OUT].ring; + + if (!in_ring || !out_ring || !dbc->ctx) { + dev_warn(dbc->dev, "Can't re-init unallocated endpoints\n"); + return -ENODEV; + } + + xhci_dbc_ring_init(in_ring); + xhci_dbc_ring_init(out_ring); + + /* set ep context enqueue, dequeue, and cycle to initial values */ + xhci_dbc_init_ep_contexts(dbc); + + return 0; +} + static struct xhci_ring * xhci_dbc_ring_alloc(struct device *dev, enum xhci_ring_type type, gfp_t flags) { @@ -885,7 +904,7 @@ static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) dev_info(dbc->dev, "DbC cable unplugged\n"); dbc->state = DS_ENABLED; xhci_dbc_flush_requests(dbc); - + xhci_dbc_reinit_ep_rings(dbc); return EVT_DISC; } @@ -895,7 +914,7 @@ static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) writel(portsc, &dbc->regs->portsc); dbc->state = DS_ENABLED; xhci_dbc_flush_requests(dbc); - + xhci_dbc_reinit_ep_rings(dbc); return EVT_DISC; }

2 months, 3 weeks

2
4
0 0

FAILED: patch "[PATCH] xhci: dbc: decouple endpoint allocation from initialization" failed to apply to 6.12-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.12-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-6.12.y git checkout FETCH_HEAD git cherry-pick -x 220a0ffde02f962c13bc752b01aa570b8c65a37b # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025091707-blip-kilometer-4c4c@gregkh' --subject-prefix 'PATCH 6.12.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 220a0ffde02f962c13bc752b01aa570b8c65a37b Mon Sep 17 00:00:00 2001 From: Mathias Nyman <mathias.nyman(a)linux.intel.com> Date: Tue, 2 Sep 2025 13:53:04 +0300 Subject: [PATCH] xhci: dbc: decouple endpoint allocation from initialization Decouple allocation of endpoint ring buffer from initialization of the buffer, and initialization of endpoint context parts from from the rest of the contexts. It allows driver to clear up and reinitialize endpoint rings after disconnect without reallocating everything. This is a prerequisite for the next patch that prevents the transfer ring from filling up with cancelled (no-op) TRBs if a debug cable is reconnected several times without transferring anything. Cc: stable(a)vger.kernel.org Fixes: dfba2174dc42 ("usb: xhci: Add DbC support in xHCI driver") Signed-off-by: Mathias Nyman <mathias.nyman(a)linux.intel.com> Link: https://lore.kernel.org/r/20250902105306.877476-2-mathias.nyman@linux.intel… Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> diff --git a/drivers/usb/host/xhci-dbgcap.c b/drivers/usb/host/xhci-dbgcap.c index 06a2edb9e86e..d0faff233e3e 100644 --- a/drivers/usb/host/xhci-dbgcap.c +++ b/drivers/usb/host/xhci-dbgcap.c @@ -101,13 +101,34 @@ static u32 xhci_dbc_populate_strings(struct dbc_str_descs *strings) return string_length; } +static void xhci_dbc_init_ep_contexts(struct xhci_dbc *dbc) +{ + struct xhci_ep_ctx *ep_ctx; + unsigned int max_burst; + dma_addr_t deq; + + max_burst = DBC_CTRL_MAXBURST(readl(&dbc->regs->control)); + + /* Populate bulk out endpoint context: */ + ep_ctx = dbc_bulkout_ctx(dbc); + deq = dbc_bulkout_enq(dbc); + ep_ctx->ep_info = 0; + ep_ctx->ep_info2 = dbc_epctx_info2(BULK_OUT_EP, 1024, max_burst); + ep_ctx->deq = cpu_to_le64(deq | dbc->ring_out->cycle_state); + + /* Populate bulk in endpoint context: */ + ep_ctx = dbc_bulkin_ctx(dbc); + deq = dbc_bulkin_enq(dbc); + ep_ctx->ep_info = 0; + ep_ctx->ep_info2 = dbc_epctx_info2(BULK_IN_EP, 1024, max_burst); + ep_ctx->deq = cpu_to_le64(deq | dbc->ring_in->cycle_state); +} + static void xhci_dbc_init_contexts(struct xhci_dbc *dbc, u32 string_length) { struct dbc_info_context *info; - struct xhci_ep_ctx *ep_ctx; u32 dev_info; - dma_addr_t deq, dma; - unsigned int max_burst; + dma_addr_t dma; if (!dbc) return; @@ -121,20 +142,8 @@ static void xhci_dbc_init_contexts(struct xhci_dbc *dbc, u32 string_length) info->serial = cpu_to_le64(dma + DBC_MAX_STRING_LENGTH * 3); info->length = cpu_to_le32(string_length); - /* Populate bulk out endpoint context: */ - ep_ctx = dbc_bulkout_ctx(dbc); - max_burst = DBC_CTRL_MAXBURST(readl(&dbc->regs->control)); - deq = dbc_bulkout_enq(dbc); - ep_ctx->ep_info = 0; - ep_ctx->ep_info2 = dbc_epctx_info2(BULK_OUT_EP, 1024, max_burst); - ep_ctx->deq = cpu_to_le64(deq | dbc->ring_out->cycle_state); - - /* Populate bulk in endpoint context: */ - ep_ctx = dbc_bulkin_ctx(dbc); - deq = dbc_bulkin_enq(dbc); - ep_ctx->ep_info = 0; - ep_ctx->ep_info2 = dbc_epctx_info2(BULK_IN_EP, 1024, max_burst); - ep_ctx->deq = cpu_to_le64(deq | dbc->ring_in->cycle_state); + /* Populate bulk in and out endpoint contexts: */ + xhci_dbc_init_ep_contexts(dbc); /* Set DbC context and info registers: */ lo_hi_writeq(dbc->ctx->dma, &dbc->regs->dccp); @@ -436,6 +445,23 @@ dbc_alloc_ctx(struct device *dev, gfp_t flags) return ctx; } +static void xhci_dbc_ring_init(struct xhci_ring *ring) +{ + struct xhci_segment *seg = ring->first_seg; + + /* clear all trbs on ring in case of old ring */ + memset(seg->trbs, 0, TRB_SEGMENT_SIZE); + + /* Only event ring does not use link TRB */ + if (ring->type != TYPE_EVENT) { + union xhci_trb *trb = &seg->trbs[TRBS_PER_SEGMENT - 1]; + + trb->link.segment_ptr = cpu_to_le64(ring->first_seg->dma); + trb->link.control = cpu_to_le32(LINK_TOGGLE | TRB_TYPE(TRB_LINK)); + } + xhci_initialize_ring_info(ring); +} + static struct xhci_ring * xhci_dbc_ring_alloc(struct device *dev, enum xhci_ring_type type, gfp_t flags) { @@ -464,15 +490,10 @@ xhci_dbc_ring_alloc(struct device *dev, enum xhci_ring_type type, gfp_t flags) seg->dma = dma; - /* Only event ring does not use link TRB */ - if (type != TYPE_EVENT) { - union xhci_trb *trb = &seg->trbs[TRBS_PER_SEGMENT - 1]; - - trb->link.segment_ptr = cpu_to_le64(dma); - trb->link.control = cpu_to_le32(LINK_TOGGLE | TRB_TYPE(TRB_LINK)); - } INIT_LIST_HEAD(&ring->td_list); - xhci_initialize_ring_info(ring); + + xhci_dbc_ring_init(ring); + return ring; dma_fail: kfree(seg);

2 months, 3 weeks

2
3
0 0

riscv: Backport request for mmap address space fixes to 6.6

by Vivian Wang

Hi, I would like to request backport of these two commits to 6.6.y: b5b4287accd702f562a49a60b10dbfaf7d40270f ("riscv: mm: Use hint address in mmap if available") 2116988d5372aec51f8c4fb85bf8e305ecda47a0 ("riscv: mm: Do not restrict mmap address based on hint") Together, they amount to disabling arch_get_mmap_end and arch_get_mmap_base for riscv. I would like to note that the first patch conflicts with commit 724103429a2d ("riscv: mm: Fixup compat arch_get_mmap_end") in the stable linux-6.6.y branch. STACK_TOP_MAX should end up as TASK_SIZE, *not* TASK_SIZE_64. Thanks, Vivian "dramforever" Wang

2 months, 3 weeks

2
1
0 0

[PATCH v2 1/3] mm/ksm: Fix incorrect KSM counter handling in mm_struct during fork

by Donet Tom

Currently, the KSM-related counters in `mm_struct`, such as `ksm_merging_pages`, `ksm_rmap_items`, and `ksm_zero_pages`, are inherited by the child process during fork. This results in inconsistent accounting. When a process uses KSM, identical pages are merged and an rmap item is created for each merged page. The `ksm_merging_pages` and `ksm_rmap_items` counters are updated accordingly. However, after a fork, these counters are copied to the child while the corresponding rmap items are not. As a result, when the child later triggers an unmerge, there are no rmap items present in the child, so the counters remain stale, leading to incorrect accounting. A similar issue exists with `ksm_zero_pages`, which maintains both a global counter and a per-process counter. During fork, the per-process counter is inherited by the child, but the global counter is not incremented. Since the child also references zero pages, the global counter should be updated as well. Otherwise, during zero-page unmerge, both the global and per-process counters are decremented, causing the global counter to become inconsistent. To fix this, ksm_merging_pages and ksm_rmap_items are reset to 0 during fork, and the global ksm_zero_pages counter is updated with the per-process ksm_zero_pages value inherited by the child. This ensures that KSM statistics remain accurate and reflect the activity of each process correctly. Fixes: 7609385337a4 ("ksm: count ksm merging pages for each process") Fixes: cb4df4cae4f2 ("ksm: count allocated ksm rmap_items for each process") Fixes: e2942062e01d ("ksm: count all zero pages placed by KSM") cc: stable(a)vger.kernel.org # v6.6 Signed-off-by: Donet Tom <donettom(a)linux.ibm.com> --- include/linux/ksm.h | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/include/linux/ksm.h b/include/linux/ksm.h index 22e67ca7cba3..067538fc4d58 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -56,8 +56,14 @@ static inline long mm_ksm_zero_pages(struct mm_struct *mm) static inline void ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm) { /* Adding mm to ksm is best effort on fork. */ - if (mm_flags_test(MMF_VM_MERGEABLE, oldmm)) + if (mm_flags_test(MMF_VM_MERGEABLE, oldmm)) { + long nr_ksm_zero_pages = atomic_long_read(&mm->ksm_zero_pages); + + mm->ksm_merging_pages = 0; + mm->ksm_rmap_items = 0; + atomic_long_add(nr_ksm_zero_pages, &ksm_zero_pages); __ksm_enter(mm); + } } static inline int ksm_execve(struct mm_struct *mm) -- 2.51.0

2 months, 3 weeks

6
9
0 0

[PATCH] media: mediatek: vcodec: Fix a reference leak in mtk_vcodec_fw_vpu_init()

by Haoxiang Li

vpu_get_plat_device() increases the reference count of the returned platform device. However, when devm_kzalloc() fails, the reference is not released, causing a reference leak. Fix this by calling put_device() on fw_pdev->dev before returning on the error path. Fixes: e25a89f743b1 ("media: mtk-vcodec: potential dereference of null pointer") Cc: stable(a)vger.kernel.org Signed-off-by: Haoxiang Li <haoxiang_li2024(a)163.com> --- .../media/platform/mediatek/vcodec/common/mtk_vcodec_fw_vpu.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_vpu.c b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_vpu.c index d7027d600208..1c94316f2d7d 100644 --- a/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_vpu.c +++ b/drivers/media/platform/mediatek/vcodec/common/mtk_vcodec_fw_vpu.c @@ -117,8 +117,10 @@ struct mtk_vcodec_fw *mtk_vcodec_fw_vpu_init(void *priv, enum mtk_vcodec_fw_use vpu_wdt_reg_handler(fw_pdev, mtk_vcodec_vpu_reset_enc_handler, priv, rst_id); fw = devm_kzalloc(&plat_dev->dev, sizeof(*fw), GFP_KERNEL); - if (!fw) + if (!fw) { + put_device(&fw_pdev->dev); return ERR_PTR(-ENOMEM); + } fw->type = VPU; fw->ops = &mtk_vcodec_vpu_msg; fw->pdev = fw_pdev; -- 2.25.1

2 months, 3 weeks

4
3
0 0

[PATCH 02/29] arm64: hugetlb: Fix flush_hugetlb_tlb_range() invalidation level

by Jia He

From: Ryan Roberts <ryan.roberts(a)arm.com> commit c910f2b65518 ("arm64/mm: Update tlb invalidation routines for FEAT_LPA2") changed the "invalidation level unknown" hint from 0 to TLBI_TTL_UNKNOWN (INT_MAX). But the fallback "unknown level" path in flush_hugetlb_tlb_range() was not updated. So as it stands, when trying to invalidate CONT_PMD_SIZE or CONT_PTE_SIZE hugetlb mappings, we will spuriously try to invalidate at level 0 on LPA2-enabled systems. Fix this so that the fallback passes TLBI_TTL_UNKNOWN, and while we are at it, explicitly use the correct stride and level for CONT_PMD_SIZE and CONT_PTE_SIZE, which should provide a minor optimization. Cc: stable(a)vger.kernel.org Fixes: c910f2b65518 ("arm64/mm: Update tlb invalidation routines for FEAT_LPA2") Reviewed-by: Anshuman Khandual <anshuman.khandual(a)arm.com> Reviewed-by: Catalin Marinas <catalin.marinas(a)arm.com> Signed-off-by: Ryan Roberts <ryan.roberts(a)arm.com> Link: https://lore.kernel.org/r/20250226120656.2400136-4-ryan.roberts@arm.com Signed-off-by: Will Deacon <will(a)kernel.org> Signed-off-by: Jia He <justin.he(a)arm.com> --- arch/arm64/include/asm/hugetlb.h | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) diff --git a/arch/arm64/include/asm/hugetlb.h b/arch/arm64/include/asm/hugetlb.h index 92a5e0879b11..eb413631edea 100644 --- a/arch/arm64/include/asm/hugetlb.h +++ b/arch/arm64/include/asm/hugetlb.h @@ -68,12 +68,22 @@ static inline void flush_hugetlb_tlb_range(struct vm_area_struct *vma, { unsigned long stride = huge_page_size(hstate_vma(vma)); - if (stride == PMD_SIZE) - __flush_tlb_range(vma, start, end, stride, false, 2); - else if (stride == PUD_SIZE) - __flush_tlb_range(vma, start, end, stride, false, 1); - else - __flush_tlb_range(vma, start, end, PAGE_SIZE, false, 0); + switch (stride) { +#ifndef __PAGETABLE_PMD_FOLDED + case PUD_SIZE: + __flush_tlb_range(vma, start, end, PUD_SIZE, false, 1); + break; +#endif + case CONT_PMD_SIZE: + case PMD_SIZE: + __flush_tlb_range(vma, start, end, PMD_SIZE, false, 2); + break; + case CONT_PTE_SIZE: + __flush_tlb_range(vma, start, end, PAGE_SIZE, false, 3); + break; + default: + __flush_tlb_range(vma, start, end, PAGE_SIZE, false, TLBI_TTL_UNKNOWN); + } } #endif /* __ASM_HUGETLB_H */ -- 2.34.1

2 months, 3 weeks

1
0
0 0

[PATCH 01/29] arm64: hugetlb: Fix huge_ptep_get_and_clear() for non-present ptes

by Jia He

From: Ryan Roberts <ryan.roberts(a)arm.com> arm64 supports multiple huge_pte sizes. Some of the sizes are covered by a single pte entry at a particular level (PMD_SIZE, PUD_SIZE), and some are covered by multiple ptes at a particular level (CONT_PTE_SIZE, CONT_PMD_SIZE). So the function has to figure out the size from the huge_pte pointer. This was previously done by walking the pgtable to determine the level and by using the PTE_CONT bit to determine the number of ptes at the level. But the PTE_CONT bit is only valid when the pte is present. For non-present pte values (e.g. markers, migration entries), the previous implementation was therefore erroneously determining the size. There is at least one known caller in core-mm, move_huge_pte(), which may call huge_ptep_get_and_clear() for a non-present pte. So we must be robust to this case. Additionally the "regular" ptep_get_and_clear() is robust to being called for non-present ptes so it makes sense to follow the behavior. Fix this by using the new sz parameter which is now provided to the function. Additionally when clearing each pte in a contig range, don't gather the access and dirty bits if the pte is not present. An alternative approach that would not require API changes would be to store the PTE_CONT bit in a spare bit in the swap entry pte for the non-present case. But it felt cleaner to follow other APIs' lead and just pass in the size. As an aside, PTE_CONT is bit 52, which corresponds to bit 40 in the swap entry offset field (layout of non-present pte). Since hugetlb is never swapped to disk, this field will only be populated for markers, which always set this bit to 0 and hwpoison swap entries, which set the offset field to a PFN; So it would only ever be 1 for a 52-bit PVA system where memory in that high half was poisoned (I think!). So in practice, this bit would almost always be zero for non-present ptes and we would only clear the first entry if it was actually a contiguous block. That's probably a less severe symptom than if it was always interpreted as 1 and cleared out potentially-present neighboring PTEs. Cc: stable(a)vger.kernel.org Fixes: 66b3923a1a0f ("arm64: hugetlb: add support for PTE contiguous bit") Reviewed-by: Catalin Marinas <catalin.marinas(a)arm.com> Signed-off-by: Ryan Roberts <ryan.roberts(a)arm.com> Link: https://lore.kernel.org/r/20250226120656.2400136-3-ryan.roberts@arm.com Signed-off-by: Will Deacon <will(a)kernel.org> Signed-off-by: Jia He <justin.he(a)arm.com> --- arch/arm64/mm/hugetlbpage.c | 53 ++++++++++++++----------------------- 1 file changed, 20 insertions(+), 33 deletions(-) diff --git a/arch/arm64/mm/hugetlbpage.c b/arch/arm64/mm/hugetlbpage.c index 7a54f8b4164b..d631d52986ec 100644 --- a/arch/arm64/mm/hugetlbpage.c +++ b/arch/arm64/mm/hugetlbpage.c @@ -121,20 +121,11 @@ static int find_num_contig(struct mm_struct *mm, unsigned long addr, static inline int num_contig_ptes(unsigned long size, size_t *pgsize) { - int contig_ptes = 0; + int contig_ptes = 1; *pgsize = size; switch (size) { -#ifndef __PAGETABLE_PMD_FOLDED - case PUD_SIZE: - if (pud_sect_supported()) - contig_ptes = 1; - break; -#endif - case PMD_SIZE: - contig_ptes = 1; - break; case CONT_PMD_SIZE: *pgsize = PMD_SIZE; contig_ptes = CONT_PMDS; @@ -143,6 +134,8 @@ static inline int num_contig_ptes(unsigned long size, size_t *pgsize) *pgsize = PAGE_SIZE; contig_ptes = CONT_PTES; break; + default: + WARN_ON(!__hugetlb_valid_size(size)); } return contig_ptes; @@ -184,24 +177,23 @@ static pte_t get_clear_contig(struct mm_struct *mm, unsigned long pgsize, unsigned long ncontig) { - pte_t orig_pte = __ptep_get(ptep); - unsigned long i; - - for (i = 0; i < ncontig; i++, addr += pgsize, ptep++) { - pte_t pte = __ptep_get_and_clear(mm, addr, ptep); - - /* - * If HW_AFDBM is enabled, then the HW could turn on - * the dirty or accessed bit for any page in the set, - * so check them all. - */ - if (pte_dirty(pte)) - orig_pte = pte_mkdirty(orig_pte); - - if (pte_young(pte)) - orig_pte = pte_mkyoung(orig_pte); + pte_t pte, tmp_pte; + bool present; + + pte = __ptep_get_and_clear(mm, addr, ptep); + present = pte_present(pte); + while (--ncontig) { + ptep++; + addr += pgsize; + tmp_pte = __ptep_get_and_clear(mm, addr, ptep); + if (present) { + if (pte_dirty(tmp_pte)) + pte = pte_mkdirty(pte); + if (pte_young(tmp_pte)) + pte = pte_mkyoung(pte); + } } - return orig_pte; + return pte; } static pte_t get_clear_contig_flush(struct mm_struct *mm, @@ -419,13 +411,8 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, { int ncontig; size_t pgsize; - pte_t orig_pte = __ptep_get(ptep); - - if (!pte_cont(orig_pte)) - return __ptep_get_and_clear(mm, addr, ptep); - - ncontig = find_num_contig(mm, addr, ptep, &pgsize); + ncontig = num_contig_ptes(sz, &pgsize); return get_clear_contig(mm, addr, ptep, pgsize, ncontig); } -- 2.34.1

2 months, 3 weeks

1
0
0 0

[PATCH 5.10] btrfs: free exchange changeset on failures

by Denis Arefev

From: Johannes Thumshirn <johannes.thumshirn(a)wdc.com> commit da5e817d9d75422eaaa05490d0b9a5e328fc1a51 upstream. Fstests runs on my VMs have show several kmemleak reports like the following. unreferenced object 0xffff88811ae59080 (size 64): comm "xfs_io", pid 12124, jiffies 4294987392 (age 6.368s) hex dump (first 32 bytes): 00 c0 1c 00 00 00 00 00 ff cf 1c 00 00 00 00 00 ................ 90 97 e5 1a 81 88 ff ff 90 97 e5 1a 81 88 ff ff ................ backtrace: [<00000000ac0176d2>] ulist_add_merge+0x60/0x150 [btrfs] [<0000000076e9f312>] set_state_bits+0x86/0xc0 [btrfs] [<0000000014fe73d6>] set_extent_bit+0x270/0x690 [btrfs] [<000000004f675208>] set_record_extent_bits+0x19/0x20 [btrfs] [<00000000b96137b1>] qgroup_reserve_data+0x274/0x310 [btrfs] [<0000000057e9dcbb>] btrfs_check_data_free_space+0x5c/0xa0 [btrfs] [<0000000019c4511d>] btrfs_delalloc_reserve_space+0x1b/0xa0 [btrfs] [<000000006d37e007>] btrfs_dio_iomap_begin+0x415/0x970 [btrfs] [<00000000fb8a74b8>] iomap_iter+0x161/0x1e0 [<0000000071dff6ff>] __iomap_dio_rw+0x1df/0x700 [<000000002567ba53>] iomap_dio_rw+0x5/0x20 [<0000000072e555f8>] btrfs_file_write_iter+0x290/0x530 [btrfs] [<000000005eb3d845>] new_sync_write+0x106/0x180 [<000000003fb505bf>] vfs_write+0x24d/0x2f0 [<000000009bb57d37>] __x64_sys_pwrite64+0x69/0xa0 [<000000003eba3fdf>] do_syscall_64+0x43/0x90 In case brtfs_qgroup_reserve_data() or btrfs_delalloc_reserve_metadata() fail the allocated extent_changeset will not be freed. So in btrfs_check_data_free_space() and btrfs_delalloc_reserve_space() free the allocated extent_changeset to get rid of the allocated memory. The issue currently only happens in the direct IO write path, but only after 65b3c08606e5 ("btrfs: fix ENOSPC failure when attempting direct IO write into NOCOW range"), and also at defrag_one_locked_target(). Every other place is always calling extent_changeset_free() even if its call to btrfs_delalloc_reserve_space() or btrfs_check_data_free_space() has failed. CC: stable(a)vger.kernel.org # 5.15+ Reviewed-by: Filipe Manana <fdmanana(a)suse.com> Signed-off-by: Johannes Thumshirn <johannes.thumshirn(a)wdc.com> Signed-off-by: David Sterba <dsterba(a)suse.com> Signed-off-by: Denis Arefev <arefev(a)swemel.ru> --- Backport fix for CVE-2021-47508 Link: https://nvd.nist.gov/vuln/detail/cve-2021-47508 --- fs/btrfs/delalloc-space.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/fs/btrfs/delalloc-space.c b/fs/btrfs/delalloc-space.c index 45b2a5ecd63a..cd12226c3220 100644 --- a/fs/btrfs/delalloc-space.c +++ b/fs/btrfs/delalloc-space.c @@ -143,10 +143,13 @@ int btrfs_check_data_free_space(struct btrfs_inode *inode, /* Use new btrfs_qgroup_reserve_data to reserve precious data space. */ ret = btrfs_qgroup_reserve_data(inode, reserved, start, len); - if (ret < 0) + if (ret < 0) { btrfs_free_reserved_data_space_noquota(fs_info, len); - else + extent_changeset_free(*reserved); + *reserved = NULL; + } else { ret = 0; + } return ret; } @@ -446,8 +449,11 @@ int btrfs_delalloc_reserve_space(struct btrfs_inode *inode, if (ret < 0) return ret; ret = btrfs_delalloc_reserve_metadata(inode, len); - if (ret < 0) + if (ret < 0) { btrfs_free_reserved_data_space(inode, *reserved, start, len); + extent_changeset_free(*reserved); + *reserved = NULL; + } return ret; } -- 2.43.0

2 months, 3 weeks

1
0
0 0

[PATCH v2] kmsan: Fix out-of-bounds access to shadow memory

by Eric Biggers

Running sha224_kunit on a KMSAN-enabled kernel results in a crash in kmsan_internal_set_shadow_origin(): BUG: unable to handle page fault for address: ffffbc3840291000 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 1810067 P4D 1810067 PUD 192d067 PMD 3c17067 PTE 0 Oops: 0000 [#1] SMP NOPTI CPU: 0 UID: 0 PID: 81 Comm: kunit_try_catch Tainted: G N 6.17.0-rc3 #10 PREEMPT(voluntary) Tainted: [N]=TEST Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.17.0-0-gb52ca86e094d-prebuilt.qemu.org 04/01/2014 RIP: 0010:kmsan_internal_set_shadow_origin+0x91/0x100 [...] Call Trace: <TASK> __msan_memset+0xee/0x1a0 sha224_final+0x9e/0x350 test_hash_buffer_overruns+0x46f/0x5f0 ? kmsan_get_shadow_origin_ptr+0x46/0xa0 ? __pfx_test_hash_buffer_overruns+0x10/0x10 kunit_try_run_case+0x198/0xa00 This occurs when memset() is called on a buffer that is not 4-byte aligned and extends to the end of a guard page, i.e. the next page is unmapped. The bug is that the loop at the end of kmsan_internal_set_shadow_origin() accesses the wrong shadow memory bytes when the address is not 4-byte aligned. Since each 4 bytes are associated with an origin, it rounds the address and size so that it can access all the origins that contain the buffer. However, when it checks the corresponding shadow bytes for a particular origin, it incorrectly uses the original unrounded shadow address. This results in reads from shadow memory beyond the end of the buffer's shadow memory, which crashes when that memory is not mapped. To fix this, correctly align the shadow address before accessing the 4 shadow bytes corresponding to each origin. Fixes: 2ef3cec44c60 ("kmsan: do not wipe out origin when doing partial unpoisoning") Cc: stable(a)vger.kernel.org Signed-off-by: Eric Biggers <ebiggers(a)kernel.org> --- v2: Added test case to kmsan_test. mm/kmsan/core.c | 10 +++++++--- mm/kmsan/kmsan_test.c | 16 ++++++++++++++++ 2 files changed, 23 insertions(+), 3 deletions(-) diff --git a/mm/kmsan/core.c b/mm/kmsan/core.c index 1ea711786c522..8bca7fece47f0 100644 --- a/mm/kmsan/core.c +++ b/mm/kmsan/core.c @@ -193,11 +193,12 @@ depot_stack_handle_t kmsan_internal_chain_origin(depot_stack_handle_t id) void kmsan_internal_set_shadow_origin(void *addr, size_t size, int b, u32 origin, bool checked) { u64 address = (u64)addr; - u32 *shadow_start, *origin_start; + void *shadow_start; + u32 *aligned_shadow, *origin_start; size_t pad = 0; KMSAN_WARN_ON(!kmsan_metadata_is_contiguous(addr, size)); shadow_start = kmsan_get_metadata(addr, KMSAN_META_SHADOW); if (!shadow_start) { @@ -212,13 +213,16 @@ void kmsan_internal_set_shadow_origin(void *addr, size_t size, int b, } return; } __memset(shadow_start, b, size); - if (!IS_ALIGNED(address, KMSAN_ORIGIN_SIZE)) { + if (IS_ALIGNED(address, KMSAN_ORIGIN_SIZE)) { + aligned_shadow = shadow_start; + } else { pad = address % KMSAN_ORIGIN_SIZE; address -= pad; + aligned_shadow = shadow_start - pad; size += pad; } size = ALIGN(size, KMSAN_ORIGIN_SIZE); origin_start = (u32 *)kmsan_get_metadata((void *)address, KMSAN_META_ORIGIN); @@ -228,11 +232,11 @@ void kmsan_internal_set_shadow_origin(void *addr, size_t size, int b, * and unconditionally overwrite the old origin slot. * If the new origin is zero, overwrite the old origin slot iff the * corresponding shadow slot is zero. */ for (int i = 0; i < size / KMSAN_ORIGIN_SIZE; i++) { - if (origin || !shadow_start[i]) + if (origin || !aligned_shadow[i]) origin_start[i] = origin; } } struct page *kmsan_vmalloc_to_page_or_null(void *vaddr) diff --git a/mm/kmsan/kmsan_test.c b/mm/kmsan/kmsan_test.c index c6c5b2bbede0c..902ec48b1e3e6 100644 --- a/mm/kmsan/kmsan_test.c +++ b/mm/kmsan/kmsan_test.c @@ -554,10 +554,25 @@ static void test_memcpy_initialized_gap(struct kunit *test) DEFINE_TEST_MEMSETXX(16) DEFINE_TEST_MEMSETXX(32) DEFINE_TEST_MEMSETXX(64) +/* Test case: ensure that KMSAN does not access shadow memory out of bounds. */ +static void test_memset_on_guarded_buffer(struct kunit *test) +{ + void *buf = vmalloc(PAGE_SIZE); + + kunit_info(test, + "memset() on ends of guarded buffer should not crash\n"); + + for (size_t size = 0; size <= 128; size++) { + memset(buf, 0xff, size); + memset(buf + PAGE_SIZE - size, 0xff, size); + } + vfree(buf); +} + static noinline void fibonacci(int *array, int size, int start) { if (start < 2 || (start == size)) return; array[start] = array[start - 1] + array[start - 2]; @@ -675,10 +690,11 @@ static struct kunit_case kmsan_test_cases[] = { KUNIT_CASE(test_memcpy_aligned_to_unaligned), KUNIT_CASE(test_memcpy_initialized_gap), KUNIT_CASE(test_memset16), KUNIT_CASE(test_memset32), KUNIT_CASE(test_memset64), + KUNIT_CASE(test_memset_on_guarded_buffer), KUNIT_CASE(test_long_origin_chain), KUNIT_CASE(test_stackdepot_roundtrip), KUNIT_CASE(test_unpoison_memory), KUNIT_CASE(test_copy_from_kernel_nofault), {}, base-commit: e59a039119c3ec241228adf12dca0dd4398104d0 -- 2.51.0

2 months, 3 weeks

2
1
0 0

Please help to cherry-pick 25daf9af0ac1 ("soc: qcom: mdt_loader: Deal with zero e_shentsize") for the 5.4/5.0/5.15/6.1 versions

by Yongqin Liu

Hi, All Please help to cherry-pick the following commit 25daf9af0ac1 ("soc: qcom: mdt_loader: Deal with zero e_shentsize") into the following branches: linux-5.4.y linux-5.10.y linux-5.15.y linux-6.1.y Which is to fix the issue caused by the following commit in the branches already: 9f9967fed9d0 ("soc: qcom: mdt_loader: Ensure we don't read past the ELF header") Just please note, for the linux-6.1.y branch the following commit needs to be cherry-picked first: 9f35ab0e53cc ("soc: qcom: mdt_loader: Fix error return values in mdt_header_valid()") before the cherry-pick of the 25daf9af0ac1 commit. # if this needs to be in a separate cherry-pick request # please let me know. -- Best Regards, Yongqin Liu --------------------------------------------------------------- #mailing list linaro-android(a)lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-android

2 months, 3 weeks

2
1
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror