October 2025 - Linux-stable-mirror

[PATCH 3/3] net: ravb: Enforce descriptor type ordering to prevent early DMA start

by Prabhakar

From: Lad Prabhakar <prabhakar.mahadev-lad.rj(a)bp.renesas.com> Ensure TX descriptor type fields are written in a safe order so the DMA engine does not begin processing a chain before all descriptors are fully initialised. For multi-descriptor transmissions the driver writes DT_FEND into the last descriptor and DT_FSTART into the first. The DMA engine starts processing when it sees DT_FSTART. If the compiler or CPU reorders the writes and publishes DT_FSTART before DT_FEND, the DMA can start early and process an incomplete chain, leading to corrupted transmissions or DMA errors. Fix this by writing DT_FEND before the dma_wmb() barrier, executing dma_wmb() immediately before DT_FSTART (or DT_FSINGLE in the single descriptor case), and then adding a wmb() after the type updates to ensure CPU-side ordering before ringing the hardware doorbell. On an RZ/G2L platform running an RT kernel, this reordering hazard was observed as TX stalls and timeouts: [ 372.968431] NETDEV WATCHDOG: end0 (ravb): transmit queue 0 timed out [ 372.968494] WARNING: CPU: 0 PID: 10 at net/sched/sch_generic.c:467 dev_watchdog+0x4a4/0x4ac [ 373.969291] ravb 11c20000.ethernet end0: transmit timed out, status 00000000, resetting... This change enforces the required ordering and prevents the DMA engine from observing DT_FSTART before the rest of the descriptor chain is valid. Fixes: 2f45d1902acf ("ravb: minimize TX data copying") Cc: stable(a)vger.kernel.org Co-developed-by: Fabrizio Castro <fabrizio.castro.jz(a)renesas.com> Signed-off-by: Fabrizio Castro <fabrizio.castro.jz(a)renesas.com> Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj(a)bp.renesas.com> --- drivers/net/ethernet/renesas/ravb_main.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index a200e205825a..2a995fa9bfff 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -2211,15 +2211,19 @@ static netdev_tx_t ravb_start_xmit(struct sk_buff *skb, struct net_device *ndev) skb_tx_timestamp(skb); } - /* Descriptor type must be set after all the above writes */ - dma_wmb(); + + /* For multi-descriptors set DT_FEND before calling dma_wmb() */ if (num_tx_desc > 1) { desc->die_dt = DT_FEND; desc--; - desc->die_dt = DT_FSTART; - } else { - desc->die_dt = DT_FSINGLE; } + + /* Descriptor type must be set after all the above writes */ + dma_wmb(); + desc->die_dt = (num_tx_desc > 1) ? DT_FSTART : DT_FSINGLE; + + /* Ensure data is written to RAM before initiating DMA transfer */ + wmb(); ravb_modify(ndev, TCCR, TCCR_TSRQ0 << q, TCCR_TSRQ0 << q); priv->cur_tx[q] += num_tx_desc; -- 2.43.0

2 weeks, 4 days

5
6
0 0

[PATCH] fs/ntfs3: fix mount failure for sparse runs in run_unpack()

by Konstantin Komarov

Some NTFS volumes failed to mount because sparse data runs were not handled correctly during runlist unpacking. The code performed arithmetic on the special SPARSE_LCN64 marker, leading to invalid LCN values and mount errors. Add an explicit check for the case described above, marking the run as sparse without applying arithmetic. Fixes: 736fc7bf5f68 ("fs: ntfs3: Fix integer overflow in run_unpack()") Cc: stable(a)vger.kernel.org Signed-off-by: Konstantin Komarov <almaz.alexandrovich(a)paragon-software.com> --- fs/ntfs3/run.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/fs/ntfs3/run.c b/fs/ntfs3/run.c index 88550085f745..5df55e4adbb1 100644 --- a/fs/ntfs3/run.c +++ b/fs/ntfs3/run.c @@ -984,8 +984,12 @@ int run_unpack(struct runs_tree *run, struct ntfs_sb_info *sbi, CLST ino, if (!dlcn) return -EINVAL; - if (check_add_overflow(prev_lcn, dlcn, &lcn)) + /* Check special combination: 0 + SPARSE_LCN64. */ + if (!prev_lcn && dlcn == SPARSE_LCN64) { + lcn = SPARSE_LCN64; + } else if (check_add_overflow(prev_lcn, dlcn, &lcn)) { return -EINVAL; + } prev_lcn = lcn; } else { /* The size of 'dlcn' can't be > 8. */ -- 2.43.0

2 weeks, 4 days

1
0
0 0

Re: Patch "selftests: mptcp: join: validate C-flag + def limit" has been added to the 5.10-stable tree

by Matthieu Baerts

Hi Greg, On 16/10/2025 15:25, gregkh(a)linuxfoundation.org wrote: > > This is a note to let you know that I've just added the patch titled > > selftests: mptcp: join: validate C-flag + def limit > > to the 5.10-stable tree which can be found at: > http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… > > The filename of the patch is: > selftests-mptcp-join-validate-c-flag-def-limit.patch > and it can be found in the queue-5.10 subdirectory. > > If you, or anyone else, feels it should not be added to the stable tree, > please let <stable(a)vger.kernel.org> know about it. I do! :) > From 008385efd05e04d8dff299382df2e8be0f91d8a0 Mon Sep 17 00:00:00 2001 > From: "Matthieu Baerts (NGI0)" <matttbe(a)kernel.org> > Date: Thu, 25 Sep 2025 12:32:37 +0200 > Subject: selftests: mptcp: join: validate C-flag + def limit > > From: Matthieu Baerts (NGI0) <matttbe(a)kernel.org> > > commit 008385efd05e04d8dff299382df2e8be0f91d8a0 upstream. > > The previous commit adds an exception for the C-flag case. The > 'mptcp_join.sh' selftest is extended to validate this case. > > In this subtest, there is a typical CDN deployment with a client where > MPTCP endpoints have been 'automatically' configured: > > - the server set net.mptcp.allow_join_initial_addr_port=0 > > - the client has multiple 'subflow' endpoints, and the default limits: > not accepting ADD_ADDRs. > > Without the parent patch, the client is not able to establish new > subflows using its 'subflow' endpoints. The parent commit fixes that. > > The 'Fixes' tag here below is the same as the one from the previous > commit: this patch here is not fixing anything wrong in the selftests, > but it validates the previous fix for an issue introduced by this commit > ID. > > Fixes: df377be38725 ("mptcp: add deny_join_id0 in mptcp_options_received") This commit has been introduced in v5.14, and not backported before. Do you mind dropping it from v5.10 please? :) Cheers, Matt -- Sponsored by the NGI0 Core fund.

2 weeks, 4 days

1
0
0 0

[PATCH v4 1/2] arm64/mm: Allow __create_pgd_mapping() to propagate pgtable_alloc() errors

by Linu Cherian

From: Chaitanya S Prakash <chaitanyas.prakash(a)arm.com> arch_add_memory() is used to hotplug memory into a system but as a part of its implementation it calls __create_pgd_mapping(), which uses pgtable_alloc() in order to build intermediate page tables. As this path was initally only used during early boot pgtable_alloc() is designed to BUG_ON() on failure. However, in the event that memory hotplug is attempted when the system's memory is extremely tight and the allocation were to fail, it would lead to panicking the system, which is not desirable. Hence update __create_pgd_mapping and all it's callers to be non void and propagate -ENOMEM on allocation failure to allow system to fail gracefully. But during early boot if there is an allocation failure, we want the system to panic, hence create a wrapper around __create_pgd_mapping() called early_create_pgd_mapping() which is designed to panic, if ret is non zero value. All the init calls are updated to use this wrapper rather than the modified __create_pgd_mapping() to restore functionality. Fixes: 4ab215061554 ("arm64: Add memory hotplug support") Cc: stable(a)vger.kernel.org Reviewed-by: Dev Jain <dev.jain(a)arm.com> Reviewed-by: Ryan Roberts <ryan.roberts(a)arm.com> Reviewed-by: Kevin Brodsky <kevin.brodsky(a)arm.com> Signed-off-by: Chaitanya S Prakash <chaitanyas.prakash(a)arm.com> Signed-off-by: Linu Cherian <linu.cherian(a)arm.com> --- Changelog: v4: * Few trivial code readability improvements arch/arm64/mm/mmu.c | 214 ++++++++++++++++++++++++++++---------------- 1 file changed, 136 insertions(+), 78 deletions(-) diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c index b8d37eb037fc..99555ebbab38 100644 --- a/arch/arm64/mm/mmu.c +++ b/arch/arm64/mm/mmu.c @@ -49,6 +49,8 @@ #define NO_CONT_MAPPINGS BIT(1) #define NO_EXEC_MAPPINGS BIT(2) /* assumes FEAT_HPDS is not used */ +#define INVALID_PHYS_ADDR (-1ULL) + DEFINE_STATIC_KEY_FALSE(arm64_ptdump_lock_key); u64 kimage_voffset __ro_after_init; @@ -194,11 +196,11 @@ static void init_pte(pte_t *ptep, unsigned long addr, unsigned long end, } while (ptep++, addr += PAGE_SIZE, addr != end); } -static void alloc_init_cont_pte(pmd_t *pmdp, unsigned long addr, - unsigned long end, phys_addr_t phys, - pgprot_t prot, - phys_addr_t (*pgtable_alloc)(enum pgtable_type), - int flags) +static int alloc_init_cont_pte(pmd_t *pmdp, unsigned long addr, + unsigned long end, phys_addr_t phys, + pgprot_t prot, + phys_addr_t (*pgtable_alloc)(enum pgtable_type), + int flags) { unsigned long next; pmd_t pmd = READ_ONCE(*pmdp); @@ -213,6 +215,8 @@ static void alloc_init_cont_pte(pmd_t *pmdp, unsigned long addr, pmdval |= PMD_TABLE_PXN; BUG_ON(!pgtable_alloc); pte_phys = pgtable_alloc(TABLE_PTE); + if (pte_phys == INVALID_PHYS_ADDR) + return -ENOMEM; ptep = pte_set_fixmap(pte_phys); init_clear_pgtable(ptep); ptep += pte_index(addr); @@ -244,11 +248,13 @@ static void alloc_init_cont_pte(pmd_t *pmdp, unsigned long addr, * walker. */ pte_clear_fixmap(); + + return 0; } -static void init_pmd(pmd_t *pmdp, unsigned long addr, unsigned long end, - phys_addr_t phys, pgprot_t prot, - phys_addr_t (*pgtable_alloc)(enum pgtable_type), int flags) +static int init_pmd(pmd_t *pmdp, unsigned long addr, unsigned long end, + phys_addr_t phys, pgprot_t prot, + phys_addr_t (*pgtable_alloc)(enum pgtable_type), int flags) { unsigned long next; @@ -269,22 +275,29 @@ static void init_pmd(pmd_t *pmdp, unsigned long addr, unsigned long end, BUG_ON(!pgattr_change_is_safe(pmd_val(old_pmd), READ_ONCE(pmd_val(*pmdp)))); } else { - alloc_init_cont_pte(pmdp, addr, next, phys, prot, - pgtable_alloc, flags); + int ret; + + ret = alloc_init_cont_pte(pmdp, addr, next, phys, prot, + pgtable_alloc, flags); + if (ret) + return ret; BUG_ON(pmd_val(old_pmd) != 0 && pmd_val(old_pmd) != READ_ONCE(pmd_val(*pmdp))); } phys += next - addr; } while (pmdp++, addr = next, addr != end); + + return 0; } -static void alloc_init_cont_pmd(pud_t *pudp, unsigned long addr, - unsigned long end, phys_addr_t phys, - pgprot_t prot, - phys_addr_t (*pgtable_alloc)(enum pgtable_type), - int flags) +static int alloc_init_cont_pmd(pud_t *pudp, unsigned long addr, + unsigned long end, phys_addr_t phys, + pgprot_t prot, + phys_addr_t (*pgtable_alloc)(enum pgtable_type), + int flags) { + int ret; unsigned long next; pud_t pud = READ_ONCE(*pudp); pmd_t *pmdp; @@ -301,6 +314,8 @@ static void alloc_init_cont_pmd(pud_t *pudp, unsigned long addr, pudval |= PUD_TABLE_PXN; BUG_ON(!pgtable_alloc); pmd_phys = pgtable_alloc(TABLE_PMD); + if (pmd_phys == INVALID_PHYS_ADDR) + return -ENOMEM; pmdp = pmd_set_fixmap(pmd_phys); init_clear_pgtable(pmdp); pmdp += pmd_index(addr); @@ -320,20 +335,26 @@ static void alloc_init_cont_pmd(pud_t *pudp, unsigned long addr, (flags & NO_CONT_MAPPINGS) == 0) __prot = __pgprot(pgprot_val(prot) | PTE_CONT); - init_pmd(pmdp, addr, next, phys, __prot, pgtable_alloc, flags); + ret = init_pmd(pmdp, addr, next, phys, __prot, pgtable_alloc, flags); + if (ret) + goto out; pmdp += pmd_index(next) - pmd_index(addr); phys += next - addr; } while (addr = next, addr != end); +out: pmd_clear_fixmap(); + + return ret; } -static void alloc_init_pud(p4d_t *p4dp, unsigned long addr, unsigned long end, - phys_addr_t phys, pgprot_t prot, - phys_addr_t (*pgtable_alloc)(enum pgtable_type), - int flags) +static int alloc_init_pud(p4d_t *p4dp, unsigned long addr, unsigned long end, + phys_addr_t phys, pgprot_t prot, + phys_addr_t (*pgtable_alloc)(enum pgtable_type), + int flags) { + int ret = 0; unsigned long next; p4d_t p4d = READ_ONCE(*p4dp); pud_t *pudp; @@ -346,6 +367,8 @@ static void alloc_init_pud(p4d_t *p4dp, unsigned long addr, unsigned long end, p4dval |= P4D_TABLE_PXN; BUG_ON(!pgtable_alloc); pud_phys = pgtable_alloc(TABLE_PUD); + if (pud_phys == INVALID_PHYS_ADDR) + return -ENOMEM; pudp = pud_set_fixmap(pud_phys); init_clear_pgtable(pudp); pudp += pud_index(addr); @@ -375,8 +398,10 @@ static void alloc_init_pud(p4d_t *p4dp, unsigned long addr, unsigned long end, BUG_ON(!pgattr_change_is_safe(pud_val(old_pud), READ_ONCE(pud_val(*pudp)))); } else { - alloc_init_cont_pmd(pudp, addr, next, phys, prot, - pgtable_alloc, flags); + ret = alloc_init_cont_pmd(pudp, addr, next, phys, prot, + pgtable_alloc, flags); + if (ret) + goto out; BUG_ON(pud_val(old_pud) != 0 && pud_val(old_pud) != READ_ONCE(pud_val(*pudp))); @@ -384,14 +409,18 @@ static void alloc_init_pud(p4d_t *p4dp, unsigned long addr, unsigned long end, phys += next - addr; } while (pudp++, addr = next, addr != end); +out: pud_clear_fixmap(); + + return ret; } -static void alloc_init_p4d(pgd_t *pgdp, unsigned long addr, unsigned long end, - phys_addr_t phys, pgprot_t prot, - phys_addr_t (*pgtable_alloc)(enum pgtable_type), - int flags) +static int alloc_init_p4d(pgd_t *pgdp, unsigned long addr, unsigned long end, + phys_addr_t phys, pgprot_t prot, + phys_addr_t (*pgtable_alloc)(enum pgtable_type), + int flags) { + int ret; unsigned long next; pgd_t pgd = READ_ONCE(*pgdp); p4d_t *p4dp; @@ -404,6 +433,8 @@ static void alloc_init_p4d(pgd_t *pgdp, unsigned long addr, unsigned long end, pgdval |= PGD_TABLE_PXN; BUG_ON(!pgtable_alloc); p4d_phys = pgtable_alloc(TABLE_P4D); + if (p4d_phys == INVALID_PHYS_ADDR) + return -ENOMEM; p4dp = p4d_set_fixmap(p4d_phys); init_clear_pgtable(p4dp); p4dp += p4d_index(addr); @@ -418,8 +449,10 @@ static void alloc_init_p4d(pgd_t *pgdp, unsigned long addr, unsigned long end, next = p4d_addr_end(addr, end); - alloc_init_pud(p4dp, addr, next, phys, prot, - pgtable_alloc, flags); + ret = alloc_init_pud(p4dp, addr, next, phys, prot, + pgtable_alloc, flags); + if (ret) + goto out; BUG_ON(p4d_val(old_p4d) != 0 && p4d_val(old_p4d) != READ_ONCE(p4d_val(*p4dp))); @@ -427,15 +460,19 @@ static void alloc_init_p4d(pgd_t *pgdp, unsigned long addr, unsigned long end, phys += next - addr; } while (p4dp++, addr = next, addr != end); +out: p4d_clear_fixmap(); + + return ret; } -static void __create_pgd_mapping_locked(pgd_t *pgdir, phys_addr_t phys, - unsigned long virt, phys_addr_t size, - pgprot_t prot, - phys_addr_t (*pgtable_alloc)(enum pgtable_type), - int flags) +static int __create_pgd_mapping_locked(pgd_t *pgdir, phys_addr_t phys, + unsigned long virt, phys_addr_t size, + pgprot_t prot, + phys_addr_t (*pgtable_alloc)(enum pgtable_type), + int flags) { + int ret; unsigned long addr, end, next; pgd_t *pgdp = pgd_offset_pgd(pgdir, virt); @@ -444,7 +481,7 @@ static void __create_pgd_mapping_locked(pgd_t *pgdir, phys_addr_t phys, * within a page, we cannot map the region as the caller expects. */ if (WARN_ON((phys ^ virt) & ~PAGE_MASK)) - return; + return -EINVAL; phys &= PAGE_MASK; addr = virt & PAGE_MASK; @@ -452,25 +489,45 @@ static void __create_pgd_mapping_locked(pgd_t *pgdir, phys_addr_t phys, do { next = pgd_addr_end(addr, end); - alloc_init_p4d(pgdp, addr, next, phys, prot, pgtable_alloc, - flags); + ret = alloc_init_p4d(pgdp, addr, next, phys, prot, pgtable_alloc, + flags); + if (ret) + return ret; phys += next - addr; } while (pgdp++, addr = next, addr != end); + + return 0; } -static void __create_pgd_mapping(pgd_t *pgdir, phys_addr_t phys, - unsigned long virt, phys_addr_t size, - pgprot_t prot, - phys_addr_t (*pgtable_alloc)(enum pgtable_type), - int flags) +static int __create_pgd_mapping(pgd_t *pgdir, phys_addr_t phys, + unsigned long virt, phys_addr_t size, + pgprot_t prot, + phys_addr_t (*pgtable_alloc)(enum pgtable_type), + int flags) { + int ret; + mutex_lock(&fixmap_lock); - __create_pgd_mapping_locked(pgdir, phys, virt, size, prot, - pgtable_alloc, flags); + ret = __create_pgd_mapping_locked(pgdir, phys, virt, size, prot, + pgtable_alloc, flags); mutex_unlock(&fixmap_lock); + + return ret; } -#define INVALID_PHYS_ADDR (-1ULL) +static void early_create_pgd_mapping(pgd_t *pgdir, phys_addr_t phys, + unsigned long virt, phys_addr_t size, + pgprot_t prot, + phys_addr_t (*pgtable_alloc)(enum pgtable_type), + int flags) +{ + int ret; + + ret = __create_pgd_mapping(pgdir, phys, virt, size, prot, pgtable_alloc, + flags); + if (ret) + panic("Failed to create page tables\n"); +} static phys_addr_t __pgd_pgtable_alloc(struct mm_struct *mm, gfp_t gfp, enum pgtable_type pgtable_type) @@ -511,21 +568,13 @@ try_pgd_pgtable_alloc_init_mm(enum pgtable_type pgtable_type, gfp_t gfp) static phys_addr_t __maybe_unused pgd_pgtable_alloc_init_mm(enum pgtable_type pgtable_type) { - phys_addr_t pa; - - pa = __pgd_pgtable_alloc(&init_mm, GFP_PGTABLE_KERNEL, pgtable_type); - BUG_ON(pa == INVALID_PHYS_ADDR); - return pa; + return __pgd_pgtable_alloc(&init_mm, GFP_PGTABLE_KERNEL, pgtable_type); } static phys_addr_t pgd_pgtable_alloc_special_mm(enum pgtable_type pgtable_type) { - phys_addr_t pa; - - pa = __pgd_pgtable_alloc(NULL, GFP_PGTABLE_KERNEL, pgtable_type); - BUG_ON(pa == INVALID_PHYS_ADDR); - return pa; + return __pgd_pgtable_alloc(NULL, GFP_PGTABLE_KERNEL, pgtable_type); } static void split_contpte(pte_t *ptep) @@ -903,8 +952,8 @@ void __init create_mapping_noalloc(phys_addr_t phys, unsigned long virt, &phys, virt); return; } - __create_pgd_mapping(init_mm.pgd, phys, virt, size, prot, NULL, - NO_CONT_MAPPINGS); + early_create_pgd_mapping(init_mm.pgd, phys, virt, size, prot, NULL, + NO_CONT_MAPPINGS); } void __init create_pgd_mapping(struct mm_struct *mm, phys_addr_t phys, @@ -918,8 +967,8 @@ void __init create_pgd_mapping(struct mm_struct *mm, phys_addr_t phys, if (page_mappings_only) flags = NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS; - __create_pgd_mapping(mm->pgd, phys, virt, size, prot, - pgd_pgtable_alloc_special_mm, flags); + early_create_pgd_mapping(mm->pgd, phys, virt, size, prot, + pgd_pgtable_alloc_special_mm, flags); } static void update_mapping_prot(phys_addr_t phys, unsigned long virt, @@ -931,8 +980,8 @@ static void update_mapping_prot(phys_addr_t phys, unsigned long virt, return; } - __create_pgd_mapping(init_mm.pgd, phys, virt, size, prot, NULL, - NO_CONT_MAPPINGS); + early_create_pgd_mapping(init_mm.pgd, phys, virt, size, prot, NULL, + NO_CONT_MAPPINGS); /* flush the TLBs after updating live kernel mappings */ flush_tlb_kernel_range(virt, virt + size); @@ -941,8 +990,8 @@ static void update_mapping_prot(phys_addr_t phys, unsigned long virt, static void __init __map_memblock(pgd_t *pgdp, phys_addr_t start, phys_addr_t end, pgprot_t prot, int flags) { - __create_pgd_mapping(pgdp, start, __phys_to_virt(start), end - start, - prot, early_pgtable_alloc, flags); + early_create_pgd_mapping(pgdp, start, __phys_to_virt(start), end - start, + prot, early_pgtable_alloc, flags); } void __init mark_linear_text_alias_ro(void) @@ -1158,6 +1207,8 @@ static int __init __kpti_install_ng_mappings(void *__unused) remap_fn = (void *)__pa_symbol(idmap_kpti_install_ng_mappings); if (!cpu) { + int ret; + alloc = __get_free_pages(GFP_ATOMIC | __GFP_ZERO, order); kpti_ng_temp_pgd = (pgd_t *)(alloc + (levels - 1) * PAGE_SIZE); kpti_ng_temp_alloc = kpti_ng_temp_pgd_pa = __pa(kpti_ng_temp_pgd); @@ -1178,9 +1229,11 @@ static int __init __kpti_install_ng_mappings(void *__unused) // covers the PTE[] page itself, the remaining entries are free // to be used as a ad-hoc fixmap. // - __create_pgd_mapping_locked(kpti_ng_temp_pgd, __pa(alloc), - KPTI_NG_TEMP_VA, PAGE_SIZE, PAGE_KERNEL, - kpti_ng_pgd_alloc, 0); + ret = __create_pgd_mapping_locked(kpti_ng_temp_pgd, __pa(alloc), + KPTI_NG_TEMP_VA, PAGE_SIZE, PAGE_KERNEL, + kpti_ng_pgd_alloc, 0); + if (ret) + panic("Failed to create page tables\n"); } cpu_install_idmap(); @@ -1233,9 +1286,9 @@ static int __init map_entry_trampoline(void) /* Map only the text into the trampoline page table */ memset(tramp_pg_dir, 0, PGD_SIZE); - __create_pgd_mapping(tramp_pg_dir, pa_start, TRAMP_VALIAS, - entry_tramp_text_size(), prot, - pgd_pgtable_alloc_init_mm, NO_BLOCK_MAPPINGS); + early_create_pgd_mapping(tramp_pg_dir, pa_start, TRAMP_VALIAS, + entry_tramp_text_size(), prot, + pgd_pgtable_alloc_init_mm, NO_BLOCK_MAPPINGS); /* Map both the text and data into the kernel page table */ for (i = 0; i < DIV_ROUND_UP(entry_tramp_text_size(), PAGE_SIZE); i++) @@ -1877,23 +1930,28 @@ int arch_add_memory(int nid, u64 start, u64 size, if (force_pte_mapping()) flags |= NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS; - __create_pgd_mapping(swapper_pg_dir, start, __phys_to_virt(start), - size, params->pgprot, pgd_pgtable_alloc_init_mm, - flags); + ret = __create_pgd_mapping(swapper_pg_dir, start, __phys_to_virt(start), + size, params->pgprot, pgd_pgtable_alloc_init_mm, + flags); + if (ret) + goto err; memblock_clear_nomap(start, size); ret = __add_pages(nid, start >> PAGE_SHIFT, size >> PAGE_SHIFT, params); if (ret) - __remove_pgd_mapping(swapper_pg_dir, - __phys_to_virt(start), size); - else { - /* Address of hotplugged memory can be smaller */ - max_pfn = max(max_pfn, PFN_UP(start + size)); - max_low_pfn = max_pfn; - } + goto err; + + /* Address of hotplugged memory can be smaller */ + max_pfn = max(max_pfn, PFN_UP(start + size)); + max_low_pfn = max_pfn; + + return 0; +err: + __remove_pgd_mapping(swapper_pg_dir, + __phys_to_virt(start), size); return ret; } -- 2.43.0

2 weeks, 4 days

2
1
0 0

[PATCH 0/2] media: i2c: imx219: Fix regressions introduced by 1x2/2x1 binning

by Jai Luthra

This series fixes a regression introduced in commit 0af46fbc333d ("media: i2c: imx219: Calculate crop rectangle dynamically") that started using vertical binning for 1920x1080 without binning horizontally, causing the captured images to be stretched in one dimension. In a subsequent patch, simplify the binning mode calculation logic as well. This is done separately, without a fixes tag, as it is cleaning up code that was introduced much later to the regression commit, and doesn't strictly require backporting. Once Sakari's metadata series with binning controls [1] is merged, this and other existing sensor drivers may be extended to support free configuration of resolution, and give the userspace freedom to bin pixels in only one dimension if required, of course, as long as it doesn't cause regressions with exsiting userspace applications. [1]: https://lore.kernel.org/all/20250825095107.1332313-44-sakari.ailus@linux.in… Signed-off-by: Jai Luthra <jai.luthra(a)ideasonboard.com> --- Dave Stevenson (1): media: i2c: imx219: Fix 1920x1080 mode to use 1:1 pixel aspect ratio Jai Luthra (1): media: i2c: imx219: Simplify imx219_get_binning() function drivers/media/i2c/imx219.c | 25 +++++++++---------------- 1 file changed, 9 insertions(+), 16 deletions(-) --- base-commit: 3a8660878839faadb4f1a6dd72c3179c1df56787 change-id: 20251016-imx219-1080p-ad3fbdce70d1 Best regards, -- Jai Luthra <jai.luthra(a)ideasonboard.com>

2 weeks, 4 days

2
2
0 0

[PATCH 6.12.y 0/4] s390/bpf: Tail call counter fixes

by Ilya Leoshkevich

Hi, As discussed in [1], here are the backports of the s390 BPF tail counter fixes. Commits 1-2 are NFC prerequisites, commits 3-4 are fixes themselves. [1] https://lore.kernel.org/stable/CA+G9fYsdErtgqKuyPfFhMS9haGKavBVCHQnipv2EeXM… Best regards, Ilya Ilya Leoshkevich (4): s390/bpf: Centralize frame offset calculations s390/bpf: Describe the frame using a struct instead of constants s390/bpf: Write back tail call counter for BPF_PSEUDO_CALL s390/bpf: Write back tail call counter for BPF_TRAMP_F_CALL_ORIG arch/s390/net/bpf_jit.h | 55 -------------- arch/s390/net/bpf_jit_comp.c | 139 ++++++++++++++++++++++------------- 2 files changed, 86 insertions(+), 108 deletions(-) delete mode 100644 arch/s390/net/bpf_jit.h -- 2.51.0

2 weeks, 4 days

1
4
0 0

[PATCH 01/12] ASoC: qdsp6: q6asm: do not sleep while atomic

by Srinivas Kandagatla

For some reason we ended up kfree between spinlock lock and unlock, which can sleep. move the kfree out of spinlock section. Fixes: a2a5d30218fd ("ASoC: qdsp6: q6asm: Add support to memory map and unmap") Cc: Stable(a)vger.kernel.org Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla(a)oss.qualcomm.com> --- sound/soc/qcom/qdsp6/q6asm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/qcom/qdsp6/q6asm.c b/sound/soc/qcom/qdsp6/q6asm.c index 06a802f9dba5..67e9ca18883c 100644 --- a/sound/soc/qcom/qdsp6/q6asm.c +++ b/sound/soc/qcom/qdsp6/q6asm.c @@ -377,9 +377,9 @@ static void q6asm_audio_client_free_buf(struct audio_client *ac, spin_lock_irqsave(&ac->lock, flags); port->num_periods = 0; + spin_unlock_irqrestore(&ac->lock, flags); kfree(port->buf); port->buf = NULL; - spin_unlock_irqrestore(&ac->lock, flags); } /** -- 2.51.0

2 weeks, 4 days

1
0
0 0

[PATCH] misc: fastrpc: Fix dma_buf object leak in fastrpc_map_lookup

by Junhao Xie

In fastrpc_map_lookup, dma_buf_get is called to obtain a reference to the dma_buf for comparison purposes. However, this reference is never released when the function returns, leading to a dma_buf memory leak. Fix this by adding dma_buf_put before returning from the function, ensuring that the temporarily acquired reference is properly released regardless of whether a matching map is found. Fixes: 9031626ade38 ("misc: fastrpc: Fix fastrpc_map_lookup operation") Cc: stable(a)kernel.org Signed-off-by: Junhao Xie <bigfoot(a)radxa.com> Tested-by: Xilin Wu <sophon(a)radxa.com> --- drivers/misc/fastrpc.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c index 621bce7e101c1..ee652ef01534a 100644 --- a/drivers/misc/fastrpc.c +++ b/drivers/misc/fastrpc.c @@ -381,6 +381,8 @@ static int fastrpc_map_lookup(struct fastrpc_user *fl, int fd, } spin_unlock(&fl->lock); + dma_buf_put(buf); + return ret; } -- 2.50.1

2 weeks, 4 days

2
1
0 0

[PATCH v2] crypto: zero initialize memory allocated via sock_kmalloc

by Shivani Agarwal

Several crypto user API contexts and requests allocated with sock_kmalloc() were left uninitialized, relying on callers to set fields explicitly. This resulted in the use of uninitialized data in certain error paths or when new fields are added in the future. The ACVP patches also contain two user-space interface files: algif_kpp.c and algif_akcipher.c. These too rely on proper initialization of their context structures. A particular issue has been observed with the newly added 'inflight' variable introduced in af_alg_ctx by commit: 67b164a871af ("crypto: af_alg - Disallow multiple in-flight AIO requests") Because the context is not memset to zero after allocation, the inflight variable has contained garbage values. As a result, af_alg_alloc_areq() has incorrectly returned -EBUSY randomly when the garbage value was interpreted as true: https://github.com/gregkh/linux/blame/master/crypto/af_alg.c#L1209 The check directly tests ctx->inflight without explicitly comparing against true/false. Since inflight is only ever set to true or false later, an uninitialized value has triggered -EBUSY failures. Zero-initializing memory allocated with sock_kmalloc() ensures inflight and other fields start in a known state, removing random issues caused by uninitialized data. Fixes: fe869cdb89c9 ("crypto: algif_hash - User-space interface for hash operations") Fixes: 5afdfd22e6ba ("crypto: algif_rng - add random number generator support") Fixes: 2d97591ef43d ("crypto: af_alg - consolidation of duplicate code") Fixes: 67b164a871af ("crypto: af_alg - Disallow multiple in-flight AIO requests") Cc: stable(a)vger.kernel.org Signed-off-by: Shivani Agarwal <shivani.agarwal(a)broadcom.com> --- Changes in v2: - Dropped algif_skcipher_export changes, The ctx->state will immediately be overwritten by crypto_skcipher_export. - No other changes. --- crypto/af_alg.c | 5 ++--- crypto/algif_hash.c | 3 +-- crypto/algif_rng.c | 3 +-- 3 files changed, 4 insertions(+), 7 deletions(-) diff --git a/crypto/af_alg.c b/crypto/af_alg.c index ca6fdcc6c54a..6c271e55f44d 100644 --- a/crypto/af_alg.c +++ b/crypto/af_alg.c @@ -1212,15 +1212,14 @@ struct af_alg_async_req *af_alg_alloc_areq(struct sock *sk, if (unlikely(!areq)) return ERR_PTR(-ENOMEM); + memset(areq, 0, areqlen); + ctx->inflight = true; areq->areqlen = areqlen; areq->sk = sk; areq->first_rsgl.sgl.sgt.sgl = areq->first_rsgl.sgl.sgl; - areq->last_rsgl = NULL; INIT_LIST_HEAD(&areq->rsgl_list); - areq->tsgl = NULL; - areq->tsgl_entries = 0; return areq; } diff --git a/crypto/algif_hash.c b/crypto/algif_hash.c index e3f1a4852737..4d3dfc60a16a 100644 --- a/crypto/algif_hash.c +++ b/crypto/algif_hash.c @@ -416,9 +416,8 @@ static int hash_accept_parent_nokey(void *private, struct sock *sk) if (!ctx) return -ENOMEM; - ctx->result = NULL; + memset(ctx, 0, len); ctx->len = len; - ctx->more = false; crypto_init_wait(&ctx->wait); ask->private = ctx; diff --git a/crypto/algif_rng.c b/crypto/algif_rng.c index 10c41adac3b1..1a86e40c8372 100644 --- a/crypto/algif_rng.c +++ b/crypto/algif_rng.c @@ -248,9 +248,8 @@ static int rng_accept_parent(void *private, struct sock *sk) if (!ctx) return -ENOMEM; + memset(ctx, 0, len); ctx->len = len; - ctx->addtl = NULL; - ctx->addtl_len = 0; /* * No seeding done at that point -- if multiple accepts are -- 2.40.4

2 weeks, 4 days

2
1
0 0

[PATCH v3] crypto: caam: Add check for kcalloc() in test_len()

by Guangshuo Li

As kcalloc() may fail, check its return value to avoid a NULL pointer dereference when passing the buffer to rng->read(). On allocation failure, log the error and return since test_len() returns void. Fixes: 2be0d806e25e ("crypto: caam - add a test for the RNG") Cc: stable(a)vger.kernel.org Signed-off-by: Guangshuo Li <lgs201920130244(a)gmail.com> --- changelog: v3: - Fix build error: test_len() returns void; return without a value. - No functional changes beyond the allocation failure path. v2: - Return on allocation failure to avoid possible NULL dereference. --- drivers/crypto/caam/caamrng.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/crypto/caam/caamrng.c b/drivers/crypto/caam/caamrng.c index b3d14a7f4dd1..0eb43c862516 100644 --- a/drivers/crypto/caam/caamrng.c +++ b/drivers/crypto/caam/caamrng.c @@ -181,7 +181,9 @@ static inline void test_len(struct hwrng *rng, size_t len, bool wait) struct device *dev = ctx->ctrldev; buf = kcalloc(CAAM_RNG_MAX_FIFO_STORE_SIZE, sizeof(u8), GFP_KERNEL); - + if (!buf) { + return; + } while (len > 0) { read_len = rng->read(rng, buf, len, wait); -- 2.43.0

2 weeks, 4 days

2
1
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror October 2025