Skip to contentSkip to conversation

5.10.253-rc1 review

PATCH In review Greg Kroah-Hartman 21 participants 570 messages Apr 13 – May 28 Page 3 of 4

Subsystems (9)
lore
Share & tools
lore
author maintainer
#295 lore raw

PATCH 277/491 net: macb: use the current queue number for stats

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paolo Valerio <[email protected]>

[ Upstream commit 72d96e4e24bbefdcfbc68bdb9341a05d8f5cb6e5 ]

There's a potential mismatch between the memory reserved for statistics and the amount of memory written.

gem_get_sset_count() correctly computes the number of stats based on the active queues, whereas gem_get_ethtool_stats() indiscriminately copies data using the maximum number of queues, and in the case the number of active queues is less than MACB_MAX_QUEUES, this results in a OOB write as observed in the KASAN splat.

==================================================================

BUG: KASAN: vmalloc-out-of-bounds in gem_get_ethtool_stats+0x54/0x78
  [macb]
Write of size 760 at addr ffff80008080b000 by task ethtool/1027

CPU: [...]

Tainted: [E]=UNSIGNED_MODULE
Hardware name: raspberrypi rpi/rpi, BIOS 2025.10 10/01/2025
Call trace:
 show_stack+0x20/0x38 (C)
 dump_stack_lvl+0x80/0xf8
 print_report+0x384/0x5e0
 kasan_report+0xa0/0xf0
 kasan_check_range+0xe8/0x190
 __asan_memcpy+0x54/0x98
 gem_get_ethtool_stats+0x54/0x78 [macb
   926c13f3af83b0c6fe64badb21ec87d5e93fcf65]
 dev_ethtool+0x1220/0x38c0
 dev_ioctl+0x4ac/0xca8
 sock_do_ioctl+0x170/0x1d8
 sock_ioctl+0x484/0x5d8
 __arm64_sys_ioctl+0x12c/0x1b8
 invoke_syscall+0xd4/0x258
 el0_svc_common.constprop.0+0xb4/0x240
 do_el0_svc+0x48/0x68
 el0_svc+0x40/0xf8
 el0t_64_sync_handler+0xa0/0xe8
 el0t_64_sync+0x1b0/0x1b8

The buggy address belongs to a 1-page vmalloc region starting at
  0xffff80008080b000 allocated at dev_ethtool+0x11f0/0x38c0
The buggy address belongs to the physical page:
page: refcount:1 mapcount:0 mapping:0000000000000000
  index:0xffff00000a333000 pfn:0xa333
flags: 0x7fffc000000000(node=0|zone=0|lastcpupid=0x1ffff)
raw: 007fffc000000000 0000000000000000 dead000000000122 0000000000000000
raw: ffff00000a333000 0000000000000000 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
 ffff80008080b080: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 ffff80008080b100: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>ffff80008080b180: 00 00 00 00 00 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
                                  ^
 ffff80008080b200: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
 ffff80008080b280: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
==================================================================

Fix it by making sure the copied size only considers the active number of queues.

---
 drivers/net/ethernet/cadence/macb_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/cadence/macb_main.c +1 -1
index f49e4e0494db3..bdbb05a95519b 100644
@@ -2786,7 +2786,7 @@ static void gem_get_ethtool_stats(struct net_device *dev,
 	spin_lock_irq(&bp->stats_lock);
 	gem_update_stats(bp);
 	memcpy(data, &bp->ethtool_stats, sizeof(u64)
-			* (GEM_STATS_LEN + QUEUE_STATS_LEN * MACB_MAX_QUEUES));
+			* (GEM_STATS_LEN + QUEUE_STATS_LEN * bp->num_queues));
 	spin_unlock_irq(&bp->stats_lock);
 }
 
Signature · 1 line
2.51.0
author maintainer
#296 lore raw

PATCH 278/491 regmap: Synchronize cache for the page selector

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Shevchenko <[email protected]>

[ Upstream commit 09e70e4f119ff650d24c96161fd2f62ac7e424b0 ]

If the selector register is represented in each page, its value according to the debugfs is stale because it gets synchronized only after the real page switch happens. Hence the regmap cache initialisation from the HW inherits outdated data in the selector register.

Synchronize cache for the page selector just in time.

Before (offset followed by hexdump, the first byte is selector):

    // Real registers
    18: 05 ff 00 00 ff 0f 00 00 f0 00 00 00
    ...
    // Virtual (per port)
    40: 05 ff 00 00 e0 e0 00 00 00 00 00 1f
    50: 00 ff 00 00 e0 e0 00 00 00 00 00 1f
    60: 01 ff 00 00 ff ff 00 00 00 00 00 00
    70: 02 ff 00 00 cf f3 00 00 00 00 00 0c
    80: 03 ff 00 00 00 00 00 00 00 00 00 ff
    90: 04 ff 00 00 ff 0f 00 00 f0 00 00 00

After:

    // Real registers
    18: 05 ff 00 00 ff 0f 00 00 f0 00 00 00
    ...
    // Virtual (per port)
    40: 00 ff 00 00 e0 e0 00 00 00 00 00 1f
    50: 01 ff 00 00 e0 e0 00 00 00 00 00 1f
    60: 02 ff 00 00 ff ff 00 00 00 00 00 00
    70: 03 ff 00 00 cf f3 00 00 00 00 00 0c
    80: 04 ff 00 00 00 00 00 00 00 00 00 ff
    90: 05 ff 00 00 ff 0f 00 00 f0 00 00 00
Fixes 6863ca622759 ("regmap: Add support for register indirect addressing.")
---
 drivers/base/regmap/regmap.c | 30 ++++++++++++++++++++++++++----
 1 file changed, 26 insertions(+), 4 deletions(-)
drivers/base/regmap/regmap.c +26 -4
index e86d069894c06..3ad37e5615ea4 100644
@@ -1587,6 +1587,7 @@ static int _regmap_select_page(struct regmap *map, unsigned int *reg,
 			       unsigned int val_num)
 {
 	void *orig_work_buf;
+	unsigned int selector_reg;
 	unsigned int win_offset;
 	unsigned int win_page;
 	bool page_chg;
@@ -1605,10 +1606,31 @@ static int _regmap_select_page(struct regmap *map, unsigned int *reg,
 			return -EINVAL;
 	}
 
-	/* It is possible to have selector register inside data window.
-	   In that case, selector register is located on every page and
-	   it needs no page switching, when accessed alone. */
+	/*
+	 * Calculate the address of the selector register in the corresponding
+	 * data window if it is located on every page.
+	 */
+	page_chg = in_range(range->selector_reg, range->window_start, range->window_len);
+	if (page_chg)
+		selector_reg = range->range_min + win_page * range->window_len +
+			       range->selector_reg - range->window_start;
+
+	/*
+	 * It is possible to have selector register inside data window.
+	 * In that case, selector register is located on every page and it
+	 * needs no page switching, when accessed alone.
+	 *
+	 * Nevertheless we should synchronize the cache values for it.
+	 * This can't be properly achieved if the selector register is
+	 * the first and the only one to be read inside the data window.
+	 * That's why we update it in that case as well.
+	 *
+	 * However, we specifically avoid updating it for the default page,
+	 * when it's overlapped with the real data window, to prevent from
+	 * infinite looping.
+	 */
 	if (val_num > 1 ||
+	    (page_chg && selector_reg != range->selector_reg) ||
 	    range->window_start + win_offset != range->selector_reg) {
 		/* Use separate work_buf during page switching */
 		orig_work_buf = map->work_buf;
@@ -1617,7 +1639,7 @@ static int _regmap_select_page(struct regmap *map, unsigned int *reg,
 		ret = _regmap_update_bits(map, range->selector_reg,
 					  range->selector_mask,
 					  win_page << range->selector_shift,
-					  &page_chg, false);
+					  NULL, false);
 
 		map->work_buf = orig_work_buf;
 
Signature · 1 line
2.53.0
author maintainer
#297 lore raw

PATCH 279/491 RDMA/rw: Fall back to direct SGE on MR pool exhaustion

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 00da250c21b074ea9494c375d0117b69e5b1d0a4 ]

When IOMMU passthrough mode is active, ib_dma_map_sgtable_attrs() produces no coalescing: each scatterlist page maps 1:1 to a DMA entry, so sgt.nents equals the raw page count. A 1 MB transfer yields 256 DMA entries. If that count exceeds the device's max_sgl_rd threshold (an optimization hint from mlx5 firmware), rdma_rw_io_needs_mr() steers the operation into the MR registration path. Each such operation consumes one or more MRs from a pool sized at max_rdma_ctxs -- roughly one MR per concurrent context. Under write-intensive workloads that issue many concurrent RDMA READs, the pool is rapidly exhausted, ib_mr_pool_get() returns NULL, and rdma_rw_init_one_mr() returns -EAGAIN. Upper layer protocols treat this as a fatal DMA mapping failure and tear down the connection.

The max_sgl_rd check is a performance optimization, not a correctness requirement: the device can handle large SGE counts via direct posting, just less efficiently than with MR registration. When the MR pool cannot satisfy a request, falling back to the direct SGE (map_wrs) path avoids the connection reset while preserving the MR optimization for the common case where pool resources are available.

Add a fallback in rdma_rw_ctx_init() so that -EAGAIN from rdma_rw_init_mr_wrs() triggers direct SGE posting instead of propagating the error. iWARP devices, which mandate MR registration for RDMA READs, and force_mr debug mode continue to treat -EAGAIN as terminal.

Fixes 00bd1439f464 ("RDMA/rw: Support threshold for registration vs scattering to local pages")
---
 drivers/infiniband/core/rw.c | 21 ++++++++++++++++++---
 1 file changed, 18 insertions(+), 3 deletions(-)
drivers/infiniband/core/rw.c +18 -3
index d479408a2cf73..58ce6c7eec4ef 100644
@@ -341,14 +341,29 @@ int rdma_rw_ctx_init(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num,
 	if (rdma_rw_io_needs_mr(qp->device, port_num, dir, sg_cnt)) {
 		ret = rdma_rw_init_mr_wrs(ctx, qp, port_num, sg, sg_cnt,
 				sg_offset, remote_addr, rkey, dir);
-	} else if (sg_cnt > 1) {
+		/*
+		 * If MR init succeeded or failed for a reason other
+		 * than pool exhaustion, that result is final.
+		 *
+		 * Pool exhaustion (-EAGAIN) from the max_sgl_rd
+		 * optimization is recoverable: fall back to
+		 * direct SGE posting. iWARP and force_mr require
+		 * MRs unconditionally, so -EAGAIN is terminal.
+		 */
+		if (ret != -EAGAIN ||
+		    rdma_protocol_iwarp(qp->device, port_num) ||
+		    unlikely(rdma_rw_force_mr))
+			goto out;
+	}
+
+	if (sg_cnt > 1)
 		ret = rdma_rw_init_map_wrs(ctx, qp, sg, sg_cnt, sg_offset,
 				remote_addr, rkey, dir);
-	} else {
+	else
 		ret = rdma_rw_init_single_wr(ctx, qp, sg, sg_offset,
 				remote_addr, rkey, dir);
-	}
 
+out:
 	if (ret < 0)
 		goto out_unmap_sg;
 	return ret;
Signature · 1 line
2.53.0
author maintainer
#298 lore raw

PATCH 280/491 scsi: scsi_transport_sas: Fix the maximum channel scanning issue

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yihang Li <[email protected]>

[ Upstream commit d71afa9deb4d413232ba16d693f7d43b321931b4 ]

After commit 37c4e72b0651 ("scsi: Fix sas_user_scan() to handle wildcard and multi-channel scans"), if the device supports multiple channels (0 to shost->max_channel), user_scan() invokes updated sas_user_scan() to perform the scan behavior for a specific transfer. However, when the user specifies shost->max_channel, it will return -EINVAL, which is not expected.

Fix and support specifying the scan shost->max_channel for scanning.

Fixes 37c4e72b0651 ("scsi: Fix sas_user_scan() to handle wildcard and multi-channel scans")
---
 drivers/scsi/scsi_transport_sas.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/scsi_transport_sas.c +1 -1
index 1eb58f8765e22..8766ec93652ca 100644
@@ -1703,7 +1703,7 @@ static int sas_user_scan(struct Scsi_Host *shost, uint channel,
 		break;
 
 	default:
-		if (channel < shost->max_channel) {
+		if (channel <= shost->max_channel) {
 			res = scsi_scan_host_selected(shost, channel, id, lun,
 						      SCSI_SCAN_MANUAL);
 		} else {
Signature · 1 line
2.53.0
author maintainer
#299 lore raw

PATCH 281/491 x86/fault: Fold mm_fault_error() into do_user_addr_fault()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Lutomirski <[email protected]>

[ Upstream commit ec352711ceba890ea3a0c182c2d49c86c1a5e30e ]

mm_fault_error() is logically just the end of do_user_addr_fault(). Combine the functions. This makes the code easier to read.

Most of the churn here is from renaming hw_error_code to error_code in do_user_addr_fault().

This makes no difference at all to the generated code (objdump -dr) as compared to changing noinline to __always_inline in the definition of mm_fault_error().

Stable-dep-of: 217c0a5c177a ("x86/efi: efi_unmap_boot_services: fix calculation of ranges_to_free size")

---
 arch/x86/mm/fault.c | 97 +++++++++++++++++++++------------------------
 1 file changed, 45 insertions(+), 52 deletions(-)
arch/x86/mm/fault.c +45 -52
index 98a5924d98b72..07983b6208f52 100644
@@ -925,40 +925,6 @@ do_sigbus(struct pt_regs *regs, unsigned long error_code, unsigned long address,
 	force_sig_fault(SIGBUS, BUS_ADRERR, (void __user *)address);
 }
 
-static noinline void
-mm_fault_error(struct pt_regs *regs, unsigned long error_code,
-	       unsigned long address, vm_fault_t fault)
-{
-	if (fatal_signal_pending(current) && !(error_code & X86_PF_USER)) {
-		no_context(regs, error_code, address, 0, 0);
-		return;
-	}
-
-	if (fault & VM_FAULT_OOM) {
-		/* Kernel mode? Handle exceptions or die: */
-		if (!(error_code & X86_PF_USER)) {
-			no_context(regs, error_code, address,
-				   SIGSEGV, SEGV_MAPERR);
-			return;
-		}
-
-		/*
-		 * We ran out of memory, call the OOM killer, and return the
-		 * userspace (which will retry the fault, or kill us if we got
-		 * oom-killed):
-		 */
-		pagefault_out_of_memory();
-	} else {
-		if (fault & (VM_FAULT_SIGBUS|VM_FAULT_HWPOISON|
-			     VM_FAULT_HWPOISON_LARGE))
-			do_sigbus(regs, error_code, address, fault);
-		else if (fault & VM_FAULT_SIGSEGV)
-			bad_area_nosemaphore(regs, error_code, address);
-		else
-			BUG();
-	}
-}
-
 static int spurious_kernel_fault_check(unsigned long error_code, pte_t *pte)
 {
 	if ((error_code & X86_PF_WRITE) && !pte_write(*pte))
@@ -1184,7 +1150,7 @@ NOKPROBE_SYMBOL(do_kern_addr_fault);
 /* Handle faults in the user portion of the address space */
 static inline
 void do_user_addr_fault(struct pt_regs *regs,
-			unsigned long hw_error_code,
+			unsigned long error_code,
 			unsigned long address)
 {
 	struct vm_area_struct *vma;
@@ -1204,8 +1170,8 @@ void do_user_addr_fault(struct pt_regs *regs,
 	 * Reserved bits are never expected to be set on
 	 * entries in the user portion of the page tables.
 	 */
-	if (unlikely(hw_error_code & X86_PF_RSVD))
-		pgtable_bad(regs, hw_error_code, address);
+	if (unlikely(error_code & X86_PF_RSVD))
+		pgtable_bad(regs, error_code, address);
 
 	/*
 	 * If SMAP is on, check for invalid kernel (supervisor) access to user
@@ -1215,10 +1181,10 @@ void do_user_addr_fault(struct pt_regs *regs,
 	 * enforcement appears to be consistent with the USER bit.
 	 */
 	if (unlikely(cpu_feature_enabled(X86_FEATURE_SMAP) &&
-		     !(hw_error_code & X86_PF_USER) &&
+		     !(error_code & X86_PF_USER) &&
 		     !(regs->flags & X86_EFLAGS_AC)))
 	{
-		bad_area_nosemaphore(regs, hw_error_code, address);
+		bad_area_nosemaphore(regs, error_code, address);
 		return;
 	}
 
@@ -1227,7 +1193,7 @@ void do_user_addr_fault(struct pt_regs *regs,
 	 * in a region with pagefaults disabled then we must not take the fault
 	 */
 	if (unlikely(faulthandler_disabled() || !mm)) {
-		bad_area_nosemaphore(regs, hw_error_code, address);
+		bad_area_nosemaphore(regs, error_code, address);
 		return;
 	}
 
@@ -1248,9 +1214,9 @@ void do_user_addr_fault(struct pt_regs *regs,
 
 	perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
 
-	if (hw_error_code & X86_PF_WRITE)
+	if (error_code & X86_PF_WRITE)
 		flags |= FAULT_FLAG_WRITE;
-	if (hw_error_code & X86_PF_INSTR)
+	if (error_code & X86_PF_INSTR)
 		flags |= FAULT_FLAG_INSTRUCTION;
 
 #ifdef CONFIG_X86_64
@@ -1266,7 +1232,7 @@ void do_user_addr_fault(struct pt_regs *regs,
 	 * to consider the PF_PK bit.
 	 */
 	if (is_vsyscall_vaddr(address)) {
-		if (emulate_vsyscall(hw_error_code, regs, address))
+		if (emulate_vsyscall(error_code, regs, address))
 			return;
 	}
 #endif
@@ -1289,7 +1255,7 @@ void do_user_addr_fault(struct pt_regs *regs,
 			 * Fault from code in kernel from
 			 * which we do not expect faults.
 			 */
-			bad_area_nosemaphore(regs, hw_error_code, address);
+			bad_area_nosemaphore(regs, error_code, address);
 			return;
 		}
 retry:
@@ -1305,17 +1271,17 @@ void do_user_addr_fault(struct pt_regs *regs,
 
 	vma = find_vma(mm, address);
 	if (unlikely(!vma)) {
-		bad_area(regs, hw_error_code, address);
+		bad_area(regs, error_code, address);
 		return;
 	}
 	if (likely(vma->vm_start <= address))
 		goto good_area;
 	if (unlikely(!(vma->vm_flags & VM_GROWSDOWN))) {
-		bad_area(regs, hw_error_code, address);
+		bad_area(regs, error_code, address);
 		return;
 	}
 	if (unlikely(expand_stack(vma, address))) {
-		bad_area(regs, hw_error_code, address);
+		bad_area(regs, error_code, address);
 		return;
 	}
 
@@ -1324,8 +1290,8 @@ void do_user_addr_fault(struct pt_regs *regs,
 	 * we can handle it..
 	 */
 good_area:
-	if (unlikely(access_error(hw_error_code, vma))) {
-		bad_area_access_error(regs, hw_error_code, address, vma);
+	if (unlikely(access_error(error_code, vma))) {
+		bad_area_access_error(regs, error_code, address, vma);
 		return;
 	}
 
@@ -1347,7 +1313,7 @@ void do_user_addr_fault(struct pt_regs *regs,
 	/* Quick path to respond to signals */
 	if (fault_signal_pending(fault, regs)) {
 		if (!user_mode(regs))
-			no_context(regs, hw_error_code, address, SIGBUS,
+			no_context(regs, error_code, address, SIGBUS,
 				   BUS_ADRERR);
 		return;
 	}
@@ -1364,11 +1330,38 @@ void do_user_addr_fault(struct pt_regs *regs,
 	}
 
 	mmap_read_unlock(mm);
-	if (unlikely(fault & VM_FAULT_ERROR)) {
-		mm_fault_error(regs, hw_error_code, address, fault);
+	if (likely(!(fault & VM_FAULT_ERROR)))
+		return;
+
+	if (fatal_signal_pending(current) && !(error_code & X86_PF_USER)) {
+		no_context(regs, error_code, address, 0, 0);
 		return;
 	}
 
+	if (fault & VM_FAULT_OOM) {
+		/* Kernel mode? Handle exceptions or die: */
+		if (!(error_code & X86_PF_USER)) {
+			no_context(regs, error_code, address,
+				   SIGSEGV, SEGV_MAPERR);
+			return;
+		}
+
+		/*
+		 * We ran out of memory, call the OOM killer, and return the
+		 * userspace (which will retry the fault, or kill us if we got
+		 * oom-killed):
+		 */
+		pagefault_out_of_memory();
+	} else {
+		if (fault & (VM_FAULT_SIGBUS|VM_FAULT_HWPOISON|
+			     VM_FAULT_HWPOISON_LARGE))
+			do_sigbus(regs, error_code, address, fault);
+		else if (fault & VM_FAULT_SIGSEGV)
+			bad_area_nosemaphore(regs, error_code, address);
+		else
+			BUG();
+	}
+
 	check_v8086_mode(regs, address, tsk);
 }
 NOKPROBE_SYMBOL(do_user_addr_fault);
Signature · 1 line
2.53.0
author maintainer
#300 lore raw

PATCH 282/491 x86/fault: Improve kernel-executing-user-memory handling

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Lutomirski <[email protected]>

[ Upstream commit 03c81ea3331658f613bb2913d33764a4e0410cbd ]

Right now, the case of the kernel trying to execute from user memory is treated more or less just like the kernel getting a page fault on a user access. In the failure path, it checks for erratum #93, tries to otherwise fix up the error, and then oopses.

If it manages to jump to the user address space, with or without SMEP, it should not try to resolve the page fault. This is an error, pure and simple. Rearrange the code so that this case is caught early, check for erratum #93, and bail out.

[ bp: Massage commit message. ]

Stable-dep-of: 217c0a5c177a ("x86/efi: efi_unmap_boot_services: fix calculation of ranges_to_free size")

---
 arch/x86/mm/fault.c | 21 ++++++++++++++++++---
 1 file changed, 18 insertions(+), 3 deletions(-)
arch/x86/mm/fault.c +18 -3
index 07983b6208f52..e06182127e1c8 100644
@@ -433,6 +433,9 @@ static int is_errata93(struct pt_regs *regs, unsigned long address)
 	    || boot_cpu_data.x86 != 0xf)
 		return 0;
 
+	if (user_mode(regs))
+		return 0;
+
 	if (address != regs->ip)
 		return 0;
 
@@ -697,9 +700,6 @@ no_context(struct pt_regs *regs, unsigned long error_code,
 	if (is_prefetch(regs, error_code, address))
 		return;
 
-	if (is_errata93(regs, address))
-		return;
-
 	/*
 	 * Buggy firmware could access regions which might page fault, try to
 	 * recover from such faults.
@@ -1162,6 +1162,21 @@ void do_user_addr_fault(struct pt_regs *regs,
 	tsk = current;
 	mm = tsk->mm;
 
+	if (unlikely((error_code & (X86_PF_USER | X86_PF_INSTR)) == X86_PF_INSTR)) {
+		/*
+		 * Whoops, this is kernel mode code trying to execute from
+		 * user memory.  Unless this is AMD erratum #93, which
+		 * corrupts RIP such that it looks like a user address,
+		 * this is unrecoverable.  Don't even try to look up the
+		 * VMA.
+		 */
+		if (is_errata93(regs, address))
+			return;
+
+		bad_area_nosemaphore(regs, error_code, address);
+		return;
+	}
+
 	/* kprobes don't want to hook the spurious faults: */
 	if (unlikely(kprobe_page_fault(regs, X86_TRAP_PF)))
 		return;
Signature · 1 line
2.53.0
author maintainer
#301 lore raw

PATCH 283/491 x86/efi: efi_unmap_boot_services: fix calculation of ranges_to_free size

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mike Rapoport (Microsoft) <[email protected]>

[ Upstream commit 217c0a5c177a3d4f7c8497950cbf5c36756e8bbb ]

ranges_to_free array should have enough room to store the entire EFI memmap plus an extra element for NULL entry.
The calculation of this array size wrongly adds 1 to the overall size instead of adding 1 to the number of elements.

Add parentheses to properly size the array.

Fixes a4b0bf6a40f3 ("x86/efi: defer freeing of boot services memory")
---
 arch/x86/platform/efi/quirks.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/platform/efi/quirks.c +1 -1
index a152d57624004..99b282d6c1302 100644
@@ -416,7 +416,7 @@ void __init efi_unmap_boot_services(void)
 	if (efi_enabled(EFI_DBG))
 		return;
 
-	sz = sizeof(*ranges_to_free) * efi.memmap.nr_map + 1;
+	sz = sizeof(*ranges_to_free) * (efi.memmap.nr_map + 1);
 	ranges_to_free = kzalloc(sz, GFP_KERNEL);
 	if (!ranges_to_free) {
 		pr_err("Failed to allocate storage for freeable EFI regions\n");
Signature · 1 line
2.53.0
author maintainer
#302 lore raw

PATCH 284/491 drm/i915/gmbus: fix spurious timeout on 512-byte burst reads

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Samasth Norway Ananda <[email protected]>

[ Upstream commit 08441f10f4dc09fdeb64529953ac308abc79dd38 ]

When reading exactly 512 bytes with burst read enabled, the extra_byte_added path breaks out of the inner do-while without decrementing len. The outer while(len) then re-enters and gmbus_wait() times out since all data has been delivered. Decrement len before the break so the outer loop terminates correctly.

Fixes d5dc0f43f268 ("drm/i915/gmbus: Enable burst read")
(cherry picked from commit 4ab0f09ee73fc853d00466682635f67c531f909c)
---
 drivers/gpu/drm/i915/display/intel_gmbus.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/display/intel_gmbus.c +3 -1
index e6b8d6dfb598e..831e99c56ecb0 100644
@@ -420,8 +420,10 @@ gmbus_xfer_read_chunk(struct drm_i915_private *dev_priv,
 
 		val = intel_de_read_fw(dev_priv, GMBUS3);
 		do {
-			if (extra_byte_added && len == 1)
+			if (extra_byte_added && len == 1) {
+				len--;
 				break;
+			}
 
 			*buf++ = val & 0xff;
 			val >>= 8;
Signature · 1 line
2.53.0
author maintainer
#303 lore raw

PATCH 285/491 ASoC: Intel: catpt: Fix the device initialization

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cezary Rojewski <[email protected]>

[ Upstream commit 5a184f1cb43a8e035251c635f5c47da5dc3e3049 ]

The DMA mask shall be coerced before any buffer allocations for the device are done. At the same time explain why DMA mask of 31 bits is used in the first place.

Cc Andy Shevchenko <[email protected]>
Fixes 7a10b66a5df9 ("ASoC: Intel: catpt: Device driver lifecycle")
---
 sound/soc/intel/catpt/device.c | 10 +++++++++-
 sound/soc/intel/catpt/dsp.c    |  3 ---
 2 files changed, 9 insertions(+), 4 deletions(-)
sound/soc/intel/catpt/device.c +9 -1
index a701799597956..db67509f051ad 100644
@@ -263,7 +263,15 @@ static int catpt_acpi_probe(struct platform_device *pdev)
 	if (IS_ERR(cdev->pci_ba))
 		return PTR_ERR(cdev->pci_ba);
 
-	/* alloc buffer for storing DRAM context during dx transitions */
+	/*
+	 * As per design HOST is responsible for preserving firmware's runtime
+	 * context during D0 -> D3 -> D0 transitions.  Addresses used for DMA
+	 * to/from HOST memory shall be outside the reserved range of 0xFFFxxxxx.
+	 */
+	ret = dma_coerce_mask_and_coherent(cdev->dev, DMA_BIT_MASK(31));
+	if (ret)
+		return ret;
+
 	cdev->dxbuf_vaddr = dmam_alloc_coherent(dev, catpt_dram_size(cdev),
 						&cdev->dxbuf_paddr, GFP_KERNEL);
 	if (!cdev->dxbuf_vaddr)
sound/soc/intel/catpt/dsp.c +0 -3
index 38a92bbc1ed56..2c67d2d35cdae 100644
@@ -125,9 +125,6 @@ int catpt_dmac_probe(struct catpt_dev *cdev)
 	dmac->dev = cdev->dev;
 	dmac->irq = cdev->irq;
 
-	ret = dma_coerce_mask_and_coherent(cdev->dev, DMA_BIT_MASK(31));
-	if (ret)
-		return ret;
 	/*
 	 * Caller is responsible for putting device in D0 to allow
 	 * for I/O and memory access before probing DW.
Signature · 1 line
2.53.0
author maintainer
#304 lore raw

PATCH 286/491 ACPICA: include/acpi/acpixf.h: Fix indentation

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

[ Upstream commit 7a9d74e7e403cb2e60d4d00c05f2f3ab2a33d0c3 ]

A bunch of the functions declared in include/acpi/acpixf.h have their name aligned a space after the '(' of e.g. the ACPI_EXTERNAL_RETURN_STATUS(acpi_status line above rather then being directly aligned after the '('.

This breaks applying patches generated from the ACPICA upstream git, remove the extra space before the function-names and all the arguments to fix this.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 include/acpi/acpixf.h | 120 +++++++++++++++++++++---------------------
 1 file changed, 60 insertions(+), 60 deletions(-)
include/acpi/acpixf.h +60 -60
index be7de305a6220..f9a86e3329715 100644
@@ -587,82 +587,82 @@ ACPI_EXTERNAL_RETURN_STATUS(acpi_status
 			    acpi_install_initialization_handler
 			    (acpi_init_handler handler, u32 function))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_install_sci_handler(acpi_sci_handler
-							  address,
-							  void *context))
-ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_remove_sci_handler(acpi_sci_handler
-							 address))
+				acpi_install_sci_handler(acpi_sci_handler
+							 address,
+							 void *context))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_install_global_event_handler
-				 (acpi_gbl_event_handler handler,
-				  void *context))
+				acpi_remove_sci_handler(acpi_sci_handler
+							address))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_install_fixed_event_handler(u32
-								  acpi_event,
-								  acpi_event_handler
-								  handler,
-								  void
-								  *context))
+				acpi_install_global_event_handler
+				(acpi_gbl_event_handler handler,
+				 void *context))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_remove_fixed_event_handler(u32 acpi_event,
+				acpi_install_fixed_event_handler(u32
+								 acpi_event,
 								 acpi_event_handler
-								 handler))
-ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_install_gpe_handler(acpi_handle
-							  gpe_device,
-							  u32 gpe_number,
-							  u32 type,
-							  acpi_gpe_handler
-							  address,
-							  void *context))
+								 handler,
+								 void
+								 *context))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_install_gpe_raw_handler(acpi_handle
-							      gpe_device,
-							      u32 gpe_number,
-							      u32 type,
-							      acpi_gpe_handler
-							      address,
-							      void *context))
+				acpi_remove_fixed_event_handler(u32 acpi_event,
+								acpi_event_handler
+								handler))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_remove_gpe_handler(acpi_handle gpe_device,
+				acpi_install_gpe_handler(acpi_handle
+							 gpe_device,
 							 u32 gpe_number,
+							 u32 type,
 							 acpi_gpe_handler
-							 address))
-ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_install_notify_handler(acpi_handle device,
-							 u32 handler_type,
-							 acpi_notify_handler
-							 handler,
+							 address,
 							 void *context))
+ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
+				acpi_install_gpe_raw_handler(acpi_handle
+							     gpe_device,
+							     u32 gpe_number,
+							     u32 type,
+							     acpi_gpe_handler
+							     address,
+							     void *context))
+ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
+				acpi_remove_gpe_handler(acpi_handle gpe_device,
+							u32 gpe_number,
+							acpi_gpe_handler
+							address))
 ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_remove_notify_handler(acpi_handle device,
+			    acpi_install_notify_handler(acpi_handle device,
 							u32 handler_type,
 							acpi_notify_handler
-							handler))
-ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_install_address_space_handler(acpi_handle
-								device,
-								acpi_adr_space_type
-								space_id,
-								acpi_adr_space_handler
-								handler,
-								acpi_adr_space_setup
-								setup,
-								void *context))
-ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_remove_address_space_handler(acpi_handle
+							handler,
+							void *context))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_remove_notify_handler(acpi_handle device,
+						       u32 handler_type,
+						       acpi_notify_handler
+						       handler))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_install_address_space_handler(acpi_handle
 							       device,
 							       acpi_adr_space_type
 							       space_id,
 							       acpi_adr_space_handler
-							       handler))
-ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_install_exception_handler
-			     (acpi_exception_handler handler))
-ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_install_interface_handler
-			     (acpi_interface_handler handler))
+							       handler,
+							       acpi_adr_space_setup
+							       setup,
+							       void *context))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_remove_address_space_handler(acpi_handle
+							      device,
+							      acpi_adr_space_type
+							      space_id,
+							      acpi_adr_space_handler
+							      handler))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_install_exception_handler
+			    (acpi_exception_handler handler))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_install_interface_handler
+			    (acpi_interface_handler handler))
 
 /*
  * Global Lock interfaces
Signature · 1 line
2.53.0
author maintainer
#305 lore raw

PATCH 287/491 ACPICA: Allow address_space_handler Install and _REG execution as 2 separate steps

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

[ Upstream commit 54c516aeb8b39eeae6450b7d8076d381568dca46 ]

ACPI-2.0 says that the EC op_region handler must be available immediately (like the standard default op_region handlers):

Quoting from the ACPI spec version 6.3: "6.5.4 _REG (Region) ...
2. OSPM must make Embedded Controller operation regions, accessed via the Embedded Controllers described in ECDT, available before executing any control method. These operation regions may become inaccessible after OSPM runs _REG(EmbeddedControl, 0)."

So the OS must probe the ECDT described EC and install the OpRegion handler before calling acpi_enable_subsystem() and acpi_initialize_objects().

This is a problem because calling acpi_install_address_space_handler() does not just install the op_region handler, it also runs the EC's _REG method. This _REG method may rely on initialization done by the _INI methods of one of the PCI / _SB root devices.

For the other early/default op_region handlers the op_region handler install and the _REG execution is split into 2 separate steps:
1. acpi_ev_install_region_handlers(), called early from acpi_load_tables()
2. acpi_ev_initialize_op_regions(), called from acpi_initialize_objects()

To fix the EC op_region issue, add 2 bew functions:
1. acpi_install_address_space_handler_no_reg()
2. acpi_execute_reg_methods()
to allow doing things in 2 steps for other op_region handlers, like the EC handler, too.

Note that the comment describing acpi_ev_install_region_handlers() even has an alinea describing this problem. Using the new methods allows users to avoid this problem.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 drivers/acpi/acpica/evxfregn.c | 92 +++++++++++++++++++++++++++++++---
 include/acpi/acpixf.h          | 10 ++++
 2 files changed, 95 insertions(+), 7 deletions(-)
drivers/acpi/acpica/evxfregn.c +85 -7
index da97fd0c6b51e..12384b3cdd6ed 100644
@@ -20,13 +20,14 @@ ACPI_MODULE_NAME("evxfregn")
 
 /*******************************************************************************
  *
- * FUNCTION:    acpi_install_address_space_handler
+ * FUNCTION:    acpi_install_address_space_handler_internal
  *
  * PARAMETERS:  device          - Handle for the device
  *              space_id        - The address space ID
  *              handler         - Address of the handler
  *              setup           - Address of the setup function
  *              context         - Value passed to the handler on each access
+ *              Run_reg         - Run _REG methods for this address space?
  *
  * RETURN:      Status
  *
@@ -37,13 +38,16 @@ ACPI_MODULE_NAME("evxfregn")
  * are executed here, and these methods can only be safely executed after
  * the default handlers have been installed and the hardware has been
  * initialized (via acpi_enable_subsystem.)
+ * To avoid this problem pass FALSE for Run_Reg and later on call
+ * acpi_execute_reg_methods() to execute _REG.
  *
  ******************************************************************************/
-acpi_status
-acpi_install_address_space_handler(acpi_handle device,
-				   acpi_adr_space_type space_id,
-				   acpi_adr_space_handler handler,
-				   acpi_adr_space_setup setup, void *context)
+static acpi_status
+acpi_install_address_space_handler_internal(acpi_handle device,
+					    acpi_adr_space_type space_id,
+					    acpi_adr_space_handler handler,
+					    acpi_adr_space_setup setup,
+					    void *context, u8 run_reg)
 {
 	struct acpi_namespace_node *node;
 	acpi_status status;
@@ -80,14 +84,40 @@ acpi_install_address_space_handler(acpi_handle device,
 
 	/* Run all _REG methods for this address space */
 
-	acpi_ev_execute_reg_methods(node, space_id, ACPI_REG_CONNECT);
+	if (run_reg) {
+		acpi_ev_execute_reg_methods(node, space_id, ACPI_REG_CONNECT);
+	}
 
 unlock_and_exit:
 	(void)acpi_ut_release_mutex(ACPI_MTX_NAMESPACE);
 	return_ACPI_STATUS(status);
 }
 
+acpi_status
+acpi_install_address_space_handler(acpi_handle device,
+				   acpi_adr_space_type space_id,
+				   acpi_adr_space_handler handler,
+				   acpi_adr_space_setup setup, void *context)
+{
+	return acpi_install_address_space_handler_internal(device, space_id,
+							   handler, setup,
+							   context, TRUE);
+}
+
 ACPI_EXPORT_SYMBOL(acpi_install_address_space_handler)
+acpi_status
+acpi_install_address_space_handler_no_reg(acpi_handle device,
+					  acpi_adr_space_type space_id,
+					  acpi_adr_space_handler handler,
+					  acpi_adr_space_setup setup,
+					  void *context)
+{
+	return acpi_install_address_space_handler_internal(device, space_id,
+							   handler, setup,
+							   context, FALSE);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_install_address_space_handler_no_reg)
 
 /*******************************************************************************
  *
@@ -226,3 +256,51 @@ acpi_remove_address_space_handler(acpi_handle device,
 }
 
 ACPI_EXPORT_SYMBOL(acpi_remove_address_space_handler)
+/*******************************************************************************
+ *
+ * FUNCTION:    acpi_execute_reg_methods
+ *
+ * PARAMETERS:  device          - Handle for the device
+ *              space_id        - The address space ID
+ *
+ * RETURN:      Status
+ *
+ * DESCRIPTION: Execute _REG for all op_regions of a given space_id.
+ *
+ ******************************************************************************/
+acpi_status
+acpi_execute_reg_methods(acpi_handle device, acpi_adr_space_type space_id)
+{
+	struct acpi_namespace_node *node;
+	acpi_status status;
+
+	ACPI_FUNCTION_TRACE(acpi_execute_reg_methods);
+
+	/* Parameter validation */
+
+	if (!device) {
+		return_ACPI_STATUS(AE_BAD_PARAMETER);
+	}
+
+	status = acpi_ut_acquire_mutex(ACPI_MTX_NAMESPACE);
+	if (ACPI_FAILURE(status)) {
+		return_ACPI_STATUS(status);
+	}
+
+	/* Convert and validate the device handle */
+
+	node = acpi_ns_validate_handle(device);
+	if (node) {
+
+		/* Run all _REG methods for this address space */
+
+		acpi_ev_execute_reg_methods(node, space_id, ACPI_REG_CONNECT);
+	} else {
+		status = AE_BAD_PARAMETER;
+	}
+
+	(void)acpi_ut_release_mutex(ACPI_MTX_NAMESPACE);
+	return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_execute_reg_methods)
include/acpi/acpixf.h +10 -0
index f9a86e3329715..890dbb8cc806a 100644
@@ -650,6 +650,16 @@ ACPI_EXTERNAL_RETURN_STATUS(acpi_status
 							       acpi_adr_space_setup
 							       setup,
 							       void *context))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_install_address_space_handler_no_reg
+			    (acpi_handle device, acpi_adr_space_type space_id,
+			     acpi_adr_space_handler handler,
+			     acpi_adr_space_setup setup,
+			     void *context))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_execute_reg_methods(acpi_handle device,
+						     acpi_adr_space_type
+						     space_id))
 ACPI_EXTERNAL_RETURN_STATUS(acpi_status
 			    acpi_remove_address_space_handler(acpi_handle
 							      device,
Signature · 1 line
2.53.0
author maintainer
#306 lore raw

PATCH 288/491 ACPI: EC: Fix EC address space handler unregistration

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

[ Upstream commit a5072078dbfaa9d70130805766dfa34bbb7bf2a7 ]

When an ECDT table is present the EC address space handler gets registered on the root node. So to unregister it properly the unregister call also must be done on the root node.

Store the ACPI handle used for the acpi_install_address_space_handler() call and use te same handle for the acpi_remove_address_space_handler() call.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 drivers/acpi/ec.c       | 4 +++-
 drivers/acpi/internal.h | 1 +
 2 files changed, 4 insertions(+), 1 deletion(-)
drivers/acpi/ec.c +3 -1
index b20206316fbe4..bf829ebc9afee 100644
@@ -1526,6 +1526,7 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device)
 			return -ENODEV;
 		}
 		set_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags);
+		ec->address_space_handler_holder = ec->handle;
 	}
 
 	if (!device)
@@ -1577,7 +1578,8 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device)
 static void ec_remove_handlers(struct acpi_ec *ec)
 {
 	if (test_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags)) {
-		if (ACPI_FAILURE(acpi_remove_address_space_handler(ec->handle,
+		if (ACPI_FAILURE(acpi_remove_address_space_handler(
+					ec->address_space_handler_holder,
 					ACPI_ADR_SPACE_EC, &acpi_ec_space_handler)))
 			pr_err("failed to remove space handler\n");
 		clear_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags);
drivers/acpi/internal.h +1 -0
index f6c929787c9e6..4edf591f8a3a5 100644
@@ -169,6 +169,7 @@ static inline void acpi_early_processor_osc(void) {}
    -------------------------------------------------------------------------- */
 struct acpi_ec {
 	acpi_handle handle;
+	acpi_handle address_space_handler_holder;
 	int gpe;
 	int irq;
 	unsigned long command_addr;
Signature · 1 line
2.53.0
author maintainer
#307 lore raw

PATCH 289/491 ACPI: EC: Fix ECDT probe ordering issues

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

[ Upstream commit ab4620f58d38206687b9f99d9d2cc1d5a2640985 ]

ACPI-2.0 says that the EC OpRegion handler must be available immediately (like the standard default OpRegion handlers):

Quoting from the ACPI spec version 6.3: "6.5.4 _REG (Region) ...
2. OSPM must make Embedded Controller operation regions, accessed via the Embedded Controllers described in ECDT, available before executing any control method. These operation regions may become inaccessible after OSPM runs _REG(EmbeddedControl, 0)."

So acpi_bus_init() calls acpi_ec_ecdt_probe(), which calls acpi_install_address_space_handler() to install the EC's OpRegion handler, early on.

This not only installs the OpRegion handler, but also calls the EC's _REG method. The _REG method call is a problem because it may rely on initialization done by the _INI methods of one of the PCI / _SB root devs, see for example: https://bugzilla.kernel.org/show_bug.cgi?id=214899 .

Generally speaking _REG methods are executed when the ACPI-device they are part of has a driver bound to it. Where as _INI methods must be executed at table load time (according to the spec). The problem here is that the early acpi_install_address_space_handler() call causes the _REG handler to run too early.

To allow fixing this the ACPICA code now allows to split the OpRegion handler installation and the executing of _REG into 2 separate steps.

This commit uses this ACPICA functionality to fix the EC probe ordering by delaying the executing of _REG for ECDT described ECs till the matching EC device in the DSDT gets parsed and acpi_ec_add() for it gets called. This moves the calling of _REG for the EC on devices with an ECDT to the same point in time where it is called on devices without an ECDT table.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 drivers/acpi/ec.c | 28 ++++++++++++++++++----------
 1 file changed, 18 insertions(+), 10 deletions(-)
drivers/acpi/ec.c +18 -10
index bf829ebc9afee..10f7e3ef58791 100644
@@ -96,6 +96,7 @@ enum {
 	EC_FLAGS_QUERY_GUARDING,	/* Guard for SCI_EVT check */
 	EC_FLAGS_EVENT_HANDLER_INSTALLED,	/* Event handler installed */
 	EC_FLAGS_EC_HANDLER_INSTALLED,	/* OpReg handler installed */
+	EC_FLAGS_EC_REG_CALLED,		/* OpReg ACPI _REG method called */
 	EC_FLAGS_QUERY_METHODS_INSTALLED, /* _Qxx handlers installed */
 	EC_FLAGS_STARTED,		/* Driver is started */
 	EC_FLAGS_STOPPED,		/* Driver is stopped */
@@ -1497,6 +1498,7 @@ static bool install_gpio_irq_event_handler(struct acpi_ec *ec)
  * ec_install_handlers - Install service callbacks and register query methods.
  * @ec: Target EC.
  * @device: ACPI device object corresponding to @ec.
+ * @call_reg: If _REG should be called to notify OpRegion availability
  *
  * Install a handler for the EC address space type unless it has been installed
  * already.  If @device is not NULL, also look for EC query methods in the
@@ -1509,7 +1511,8 @@ static bool install_gpio_irq_event_handler(struct acpi_ec *ec)
  * -EPROBE_DEFER if GPIO IRQ acquisition needs to be deferred,
  * or 0 (success) otherwise.
  */
-static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device)
+static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
+			       bool call_reg)
 {
 	acpi_status status;
 
@@ -1517,10 +1520,10 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device)
 
 	if (!test_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags)) {
 		acpi_ec_enter_noirq(ec);
-		status = acpi_install_address_space_handler(ec->handle,
-							    ACPI_ADR_SPACE_EC,
-							    &acpi_ec_space_handler,
-							    NULL, ec);
+		status = acpi_install_address_space_handler_no_reg(ec->handle,
+								   ACPI_ADR_SPACE_EC,
+								   &acpi_ec_space_handler,
+								   NULL, ec);
 		if (ACPI_FAILURE(status)) {
 			acpi_ec_stop(ec, false);
 			return -ENODEV;
@@ -1529,6 +1532,11 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device)
 		ec->address_space_handler_holder = ec->handle;
 	}
 
+	if (call_reg && !test_bit(EC_FLAGS_EC_REG_CALLED, &ec->flags)) {
+		acpi_execute_reg_methods(ec->handle, ACPI_ADR_SPACE_EC);
+		set_bit(EC_FLAGS_EC_REG_CALLED, &ec->flags);
+	}
+
 	if (!device)
 		return 0;
 
@@ -1615,11 +1623,11 @@ static void ec_remove_handlers(struct acpi_ec *ec)
 	}
 }
 
-static int acpi_ec_setup(struct acpi_ec *ec, struct acpi_device *device)
+static int acpi_ec_setup(struct acpi_ec *ec, struct acpi_device *device, bool call_reg)
 {
 	int ret;
 
-	ret = ec_install_handlers(ec, device);
+	ret = ec_install_handlers(ec, device, call_reg);
 	if (ret)
 		return ret;
 
@@ -1681,7 +1689,7 @@ static int acpi_ec_add(struct acpi_device *device)
 		}
 	}
 
-	ret = acpi_ec_setup(ec, device);
+	ret = acpi_ec_setup(ec, device, true);
 	if (ret)
 		goto err;
 
@@ -1801,7 +1809,7 @@ void __init acpi_ec_dsdt_probe(void)
 	 * At this point, the GPE is not fully initialized, so do not to
 	 * handle the events.
 	 */
-	ret = acpi_ec_setup(ec, NULL);
+	ret = acpi_ec_setup(ec, NULL, true);
 	if (ret) {
 		acpi_ec_free(ec);
 		return;
@@ -1965,7 +1973,7 @@ void __init acpi_ec_ecdt_probe(void)
 	 * At this point, the namespace is not initialized, so do not find
 	 * the namespace objects, or handle the events.
 	 */
-	ret = acpi_ec_setup(ec, NULL);
+	ret = acpi_ec_setup(ec, NULL, false);
 	if (ret) {
 		acpi_ec_free(ec);
 		goto out;
Signature · 1 line
2.53.0
author maintainer
#308 lore raw

PATCH 290/491 ACPI: EC: Install address space handler at the namespace root

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafael J. Wysocki <[email protected]>

[ Upstream commit 60fa6ae6e6d09e377fce6f8d9b6f6a4d88769f63 ]

It is reported that _DSM evaluation fails in ucsi_acpi_dsm() on Lenovo IdeaPad Pro 5 due to a missing address space handler for the EC address space:

ACPI Error: No handler for Region [ECSI] (000000007b8176ee) [EmbeddedControl] (20230628/evregion-130)

This happens because if there is no ECDT, the EC driver only registers the EC address space handler for operation regions defined in the EC device scope of the ACPI namespace while the operation region being accessed by the _DSM in question is located beyond that scope.

To address this, modify the ACPI EC driver to install the EC address space handler at the root of the ACPI namespace for the first EC that can be found regardless of whether or not an ECDT is present.

Note that this change is consistent with some examples in the ACPI specification in which EC operation regions located outside the EC device scope are used (for example, see Section 9.17.15 in ACPI 6.5), so the current behavior of the EC driver is arguably questionable.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 drivers/acpi/ec.c       | 25 ++++++++++++++++---------
 drivers/acpi/internal.h |  1 -
 2 files changed, 16 insertions(+), 10 deletions(-)
drivers/acpi/ec.c +16 -9
index 10f7e3ef58791..1d7e7e47ea0e4 100644
@@ -1514,13 +1514,14 @@ static bool install_gpio_irq_event_handler(struct acpi_ec *ec)
 static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
 			       bool call_reg)
 {
+	acpi_handle scope_handle = ec == first_ec ? ACPI_ROOT_OBJECT : ec->handle;
 	acpi_status status;
 
 	acpi_ec_start(ec, false);
 
 	if (!test_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags)) {
 		acpi_ec_enter_noirq(ec);
-		status = acpi_install_address_space_handler_no_reg(ec->handle,
+		status = acpi_install_address_space_handler_no_reg(scope_handle,
 								   ACPI_ADR_SPACE_EC,
 								   &acpi_ec_space_handler,
 								   NULL, ec);
@@ -1529,11 +1530,10 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
 			return -ENODEV;
 		}
 		set_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags);
-		ec->address_space_handler_holder = ec->handle;
 	}
 
 	if (call_reg && !test_bit(EC_FLAGS_EC_REG_CALLED, &ec->flags)) {
-		acpi_execute_reg_methods(ec->handle, ACPI_ADR_SPACE_EC);
+		acpi_execute_reg_methods(scope_handle, ACPI_ADR_SPACE_EC);
 		set_bit(EC_FLAGS_EC_REG_CALLED, &ec->flags);
 	}
 
@@ -1585,10 +1585,13 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
 
 static void ec_remove_handlers(struct acpi_ec *ec)
 {
+	acpi_handle scope_handle = ec == first_ec ? ACPI_ROOT_OBJECT : ec->handle;
+
 	if (test_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags)) {
 		if (ACPI_FAILURE(acpi_remove_address_space_handler(
-					ec->address_space_handler_holder,
-					ACPI_ADR_SPACE_EC, &acpi_ec_space_handler)))
+						scope_handle,
+						ACPI_ADR_SPACE_EC,
+						&acpi_ec_space_handler)))
 			pr_err("failed to remove space handler\n");
 		clear_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags);
 	}
@@ -1627,14 +1630,18 @@ static int acpi_ec_setup(struct acpi_ec *ec, struct acpi_device *device, bool ca
 {
 	int ret;
 
-	ret = ec_install_handlers(ec, device, call_reg);
-	if (ret)
-		return ret;
-
 	/* First EC capable of handling transactions */
 	if (!first_ec)
 		first_ec = ec;
 
+	ret = ec_install_handlers(ec, device, call_reg);
+	if (ret) {
+		if (ec == first_ec)
+			first_ec = NULL;
+
+		return ret;
+	}
+
 	pr_info("EC_CMD/EC_SC=0x%lx, EC_DATA=0x%lx\n", ec->command_addr,
 		ec->data_addr);
 
drivers/acpi/internal.h +0 -1
index 4edf591f8a3a5..f6c929787c9e6 100644
@@ -169,7 +169,6 @@ static inline void acpi_early_processor_osc(void) {}
    -------------------------------------------------------------------------- */
 struct acpi_ec {
 	acpi_handle handle;
-	acpi_handle address_space_handler_holder;
 	int gpe;
 	int irq;
 	unsigned long command_addr;
Signature · 1 line
2.53.0
#309 lore raw

Re: [PATCH 5.10 290/491] ACPI: EC: Install address space handler at the namespace root

Quoted 7 lines · Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafael J. Wysocki <[email protected]>

[ Upstream commit 60fa6ae6e6d09e377fce6f8d9b6f6a4d88769f63 ]

It appears that this can cause regressions, which were fixed by commits 71bf41b8e913 ("ACPI: EC: Evaluate _REG outside the EC scope more carefully") and 0e6b6dedf168 ("ACPI: EC: Evaluate orphan _REG under EC device") upstream. I didn't check whether those would apply and build cleanly here, but if not then this may have to be dropped.

Ben.

Quoted 120 lines

It is reported that _DSM evaluation fails in ucsi_acpi_dsm() on Lenovo IdeaPad Pro 5 due to a missing address space handler for the EC address space:

ACPI Error: No handler for Region [ECSI] (000000007b8176ee) [EmbeddedControl] (20230628/evregion-130)

This happens because if there is no ECDT, the EC driver only registers the EC address space handler for operation regions defined in the EC device scope of the ACPI namespace while the operation region being accessed by the _DSM in question is located beyond that scope.

To address this, modify the ACPI EC driver to install the EC address space handler at the root of the ACPI namespace for the first EC that can be found regardless of whether or not an ECDT is present.

Note that this change is consistent with some examples in the ACPI specification in which EC operation regions located outside the EC device scope are used (for example, see Section 9.17.15 in ACPI 6.5), so the current behavior of the EC driver is arguably questionable.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 drivers/acpi/ec.c       | 25 ++++++++++++++++---------
 drivers/acpi/internal.h |  1 -
 2 files changed, 16 insertions(+), 10 deletions(-)
drivers/acpi/ec.c +16 -9
index 10f7e3ef58791..1d7e7e47ea0e4 100644
@@ -1514,13 +1514,14 @@ static bool install_gpio_irq_event_handler(struct acpi_ec *ec)
 static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
 			       bool call_reg)
 {
+	acpi_handle scope_handle = ec == first_ec ? ACPI_ROOT_OBJECT : ec->handle;
 	acpi_status status;
 
 	acpi_ec_start(ec, false);
 
 	if (!test_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags)) {
 		acpi_ec_enter_noirq(ec);
-		status = acpi_install_address_space_handler_no_reg(ec->handle,
+		status = acpi_install_address_space_handler_no_reg(scope_handle,
 								   ACPI_ADR_SPACE_EC,
 								   &acpi_ec_space_handler,
 								   NULL, ec);
@@ -1529,11 +1530,10 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
 			return -ENODEV;
 		}
 		set_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags);
-		ec->address_space_handler_holder = ec->handle;
 	}
 
 	if (call_reg && !test_bit(EC_FLAGS_EC_REG_CALLED, &ec->flags)) {
-		acpi_execute_reg_methods(ec->handle, ACPI_ADR_SPACE_EC);
+		acpi_execute_reg_methods(scope_handle, ACPI_ADR_SPACE_EC);
 		set_bit(EC_FLAGS_EC_REG_CALLED, &ec->flags);
 	}
 
@@ -1585,10 +1585,13 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
 
 static void ec_remove_handlers(struct acpi_ec *ec)
 {
+	acpi_handle scope_handle = ec == first_ec ? ACPI_ROOT_OBJECT : ec->handle;
+
 	if (test_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags)) {
 		if (ACPI_FAILURE(acpi_remove_address_space_handler(
-					ec->address_space_handler_holder,
-					ACPI_ADR_SPACE_EC, &acpi_ec_space_handler)))
+						scope_handle,
+						ACPI_ADR_SPACE_EC,
+						&acpi_ec_space_handler)))
 			pr_err("failed to remove space handler\n");
 		clear_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags);
 	}
@@ -1627,14 +1630,18 @@ static int acpi_ec_setup(struct acpi_ec *ec, struct acpi_device *device, bool ca
 {
 	int ret;
 
-	ret = ec_install_handlers(ec, device, call_reg);
-	if (ret)
-		return ret;
-
 	/* First EC capable of handling transactions */
 	if (!first_ec)
 		first_ec = ec;
 
+	ret = ec_install_handlers(ec, device, call_reg);
+	if (ret) {
+		if (ec == first_ec)
+			first_ec = NULL;
+
+		return ret;
+	}
+
 	pr_info("EC_CMD/EC_SC=0x%lx, EC_DATA=0x%lx\n", ec->command_addr,
 		ec->data_addr);
 
drivers/acpi/internal.h +0 -1
index 4edf591f8a3a5..f6c929787c9e6 100644
@@ -169,7 +169,6 @@ static inline void acpi_early_processor_osc(void) {}
    -------------------------------------------------------------------------- */
 struct acpi_ec {
 	acpi_handle handle;
-	acpi_handle address_space_handler_holder;
 	int gpe;
 	int irq;
 	unsigned long command_addr;
Signature · 2 lines
Ben Hutchings
When in doubt, use brute force. - Ken Thompson
maintainer
#310 lore raw

Re: [PATCH 5.10 290/491] ACPI: EC: Install address space handler at the namespace root

Quoted 4 lines · Ben Hutchings

It appears that this can cause regressions, which were fixed by commits 71bf41b8e913 ("ACPI: EC: Evaluate _REG outside the EC scope more carefully") and 0e6b6dedf168 ("ACPI: EC: Evaluate orphan _REG under EC device") upstream.

Dropped from the 5.15 and 5.10 queues along with the following dependencies:
- "ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()" (5.15, 5.10)
- "ACPI: EC: Evaluate orphan _REG under EC device" (5.15)

Thanks for catching this.

author maintainer
#311 lore raw

PATCH 291/491 ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Weiming Shi <[email protected]>

[ Upstream commit f6484cadbcaf26b5844b51bd7307a663dda48ef6 ]

When ec_install_handlers() returns -EPROBE_DEFER on reduced-hardware platforms, it has already started the EC and installed the address space handler with the struct acpi_ec pointer as handler context. However, acpi_ec_setup() propagates the error without any cleanup.

The caller acpi_ec_add() then frees the struct acpi_ec for non-boot instances, leaving a dangling handler context in ACPICA.

Any subsequent AML evaluation that accesses an EC OpRegion field dispatches into acpi_ec_space_handler() with the freed pointer, causing a use-after-free:

 BUG: KASAN: slab-use-after-free in mutex_lock (kernel/locking/mutex.c:289)
 Write of size 8 at addr ffff88800721de38 by task init/1
 Call Trace:
  <TASK>
  mutex_lock (kernel/locking/mutex.c:289)
  acpi_ec_space_handler (drivers/acpi/ec.c:1362)
  acpi_ev_address_space_dispatch (drivers/acpi/acpica/evregion.c:293)
  acpi_ex_access_region (drivers/acpi/acpica/exfldio.c:246)
  acpi_ex_field_datum_io (drivers/acpi/acpica/exfldio.c:509)
  acpi_ex_extract_from_field (drivers/acpi/acpica/exfldio.c:700)
  acpi_ex_read_data_from_field (drivers/acpi/acpica/exfield.c:327)
  acpi_ex_resolve_node_to_value (drivers/acpi/acpica/exresolv.c:392)
  </TASK>

 Allocated by task 1:
  acpi_ec_alloc (drivers/acpi/ec.c:1424)
  acpi_ec_add (drivers/acpi/ec.c:1692)

 Freed by task 1:
  kfree (mm/slub.c:6876)
  acpi_ec_add (drivers/acpi/ec.c:1751)

The bug triggers on reduced-hardware EC platforms (ec->gpe < 0) when the GPIO IRQ provider defers probing. Once the stale handler exists, any unprivileged sysfs read that causes AML to touch an EC OpRegion (battery, thermal, backlight) exercises the dangling pointer.

Fix this by calling ec_remove_handlers() in the error path of acpi_ec_setup() before clearing first_ec. ec_remove_handlers() checks each EC_FLAGS_* bit before acting, so it is safe to call regardless of how far ec_install_handlers() progressed:

-ENODEV (handler not installed): only calls acpi_ec_stop() -EPROBE_DEFER (handler installed): removes handler, stops EC

Fixes 03e9a0e05739 ("ACPI: EC: Consolidate event handler installation code")
---
 drivers/acpi/ec.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/acpi/ec.c +2 -0
index 1d7e7e47ea0e4..aaf0add0dc081 100644
@@ -1636,6 +1636,8 @@ static int acpi_ec_setup(struct acpi_ec *ec, struct acpi_device *device, bool ca
 
 	ret = ec_install_handlers(ec, device, call_reg);
 	if (ret) {
+		ec_remove_handlers(ec);
+
 		if (ec == first_ec)
 			first_ec = NULL;
 
Signature · 1 line
2.53.0
author maintainer
#312 lore raw

PATCH 292/491 hwmon: (adm1177) fix sysfs ABI violation and current unit conversion

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sanman Pradhan <[email protected]>

[ Upstream commit bf08749a6abb6d1959bfdc0edc32c640df407558 ]

The adm1177 driver exposes the current alert threshold through hwmon_curr_max_alarm. This violates the hwmon sysfs ABI, where *_alarm attributes are read-only status flags and writable thresholds must use currN_max.

The driver also stores the threshold internally in microamps, while currN_max is defined in milliamps. Convert the threshold accordingly on both the read and write paths.

Widen the cached threshold and related calculations to 64 bits so that small shunt resistor values do not cause truncation or overflow. Also use 64-bit arithmetic for the mA/uA conversions, clamp writes to the range the hardware can represent, and propagate failures from adm1177_write_alert_thr() instead of silently ignoring them.

Update the hwmon documentation to reflect the attribute rename and the correct units returned by the driver.

Fixes 09b08ac9e8d5 ("hwmon: (adm1177) Add ADM1177 Hot Swap Controller and Digital Power Monitor driver")
---
 Documentation/hwmon/adm1177.rst |  8 ++---
 drivers/hwmon/adm1177.c         | 54 +++++++++++++++++++--------------
 2 files changed, 35 insertions(+), 27 deletions(-)
Documentation/hwmon/adm1177.rst +4 -4
index 471be1e98d6f6..fc0c3390c302a 100644
@@ -26,10 +26,10 @@ devices explicitly. Please see :doc:`/i2c/instantiating-devices` for details.
 Sysfs entries
 -------------
 
-The following attributes are supported. Current maxim attribute
+The following attributes are supported. Current maximum attribute
 is read-write, all other attributes are read-only.
 
-in0_input		Measured voltage in microvolts.
+in0_input		Measured voltage in millivolts.
 
-curr1_input		Measured current in microamperes.
-curr1_max_alarm		Overcurrent alarm in microamperes.
+curr1_input		Measured current in milliamperes.
+curr1_max		Overcurrent shutdown threshold in milliamperes.
drivers/hwmon/adm1177.c +31 -23

111 lines not on this page · raw message

Signature · 1 line
2.53.0
author maintainer
#313 lore raw

PATCH 293/491 sysctl: fix uninitialized variable in proc_do_large_bitmap

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marc Buerg <[email protected]>

[ Upstream commit f63a9df7e3f9f842945d292a19d9938924f066f9 ]

proc_do_large_bitmap() does not initialize variable c, which is expected to be set to a trailing character by proc_get_long().

However, proc_get_long() only sets c when the input buffer contains a trailing character after the parsed value.

If c is not initialized it may happen to contain a '-'. If this is the case proc_do_large_bitmap() expects to be able to parse a second part of the input buffer. If there is no second part an unjustified -EINVAL will be returned.

Initialize c to 0 to prevent returning -EINVAL on valid input.

Fixes 9f977fb7ae9d ("sysctl: add proc_do_large_bitmap")
---
 kernel/sysctl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/sysctl.c +1 -1
index abe0f16d53641..c2870c9cae14a 100644
@@ -1549,7 +1549,7 @@ int proc_do_large_bitmap(struct ctl_table *table, int write,
 	unsigned long bitmap_len = table->maxlen;
 	unsigned long *bitmap = *(unsigned long **) table->data;
 	unsigned long *tmp_bitmap = NULL;
-	char tr_a[] = { '-', ',', '\n' }, tr_b[] = { ',', '\n', 0 }, c;
+	char tr_a[] = { '-', ',', '\n' }, tr_b[] = { ',', '\n', 0 }, c = 0;
 
 	if (!bitmap || !bitmap_len || !left || (*ppos && !write)) {
 		*lenp = 0;
Signature · 1 line
2.53.0
author maintainer
#314 lore raw

PATCH 294/491 spi: spi-fsl-lpspi: fix teardown order issue (UAF)

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marc Kleine-Budde <[email protected]>

[ Upstream commit b341c1176f2e001b3adf0b47154fc31589f7410e ]

There is a teardown order issue in the driver. The SPI controller is registered using devm_spi_register_controller(), which delays unregistration of the SPI controller until after the fsl_lpspi_remove() function returns.

As the fsl_lpspi_remove() function synchronously tears down the DMA channels, a running SPI transfer triggers the following NULL pointer dereference due to use after free:

| fsl_lpspi 42550000.spi: I/O Error in DMA RX
| Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000 [...]

| Call trace:
|  fsl_lpspi_dma_transfer+0x260/0x340 [spi_fsl_lpspi]
|  fsl_lpspi_transfer_one+0x198/0x448 [spi_fsl_lpspi]
|  spi_transfer_one_message+0x49c/0x7c8
|  __spi_pump_transfer_message+0x120/0x420
|  __spi_sync+0x2c4/0x520
|  spi_sync+0x34/0x60
|  spidev_message+0x20c/0x378 [spidev]
|  spidev_ioctl+0x398/0x750 [spidev]

[...]

Switch from devm_spi_register_controller() to spi_register_controller() in fsl_lpspi_probe() and add the corresponding spi_unregister_controller() in fsl_lpspi_remove().

---
 drivers/spi/spi-fsl-lpspi.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/spi/spi-fsl-lpspi.c +2 -1
@@ -914,7 +914,7 @@ static int fsl_lpspi_probe(struct platfo
 		enable_irq(irq);
 	}
 
-	ret = devm_spi_register_controller(&pdev->dev, controller);
+	ret = spi_register_controller(controller);
 	if (ret < 0) {
 		dev_err_probe(&pdev->dev, ret, "spi_register_controller error\n");
 		goto free_dma;
@@ -943,6 +943,7 @@ static int fsl_lpspi_remove(struct platf
 	struct fsl_lpspi_data *fsl_lpspi =
 				spi_controller_get_devdata(controller);
 
+	spi_unregister_controller(controller);
 	fsl_lpspi_dma_exit(controller);
 
 	pm_runtime_dont_use_autosuspend(fsl_lpspi->dev);
author maintainer
#315 lore raw

PATCH 295/491 s390/barrier: Make array_index_mask_nospec() __always_inline

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vasily Gorbik <[email protected]>

commit c5c0a268b38adffbb2e70e6957017537ff54c157 upstream.

Mark array_index_mask_nospec() as __always_inline to guarantee the mitigation is emitted inline regardless of compiler inlining decisions.

Fixes e2dd833389cc ("s390: add optimized array_index_mask_nospec")
---
 arch/s390/include/asm/barrier.h |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/s390/include/asm/barrier.h +2 -2
@@ -56,8 +56,8 @@ do {									\
  * @size: number of elements in array
  */
 #define array_index_mask_nospec array_index_mask_nospec
-static inline unsigned long array_index_mask_nospec(unsigned long index,
-						    unsigned long size)
+static __always_inline unsigned long array_index_mask_nospec(unsigned long index,
+							     unsigned long size)
 {
 	unsigned long mask;
 
author maintainer
#316 lore raw

PATCH 296/491 can: gw: fix OOB heap access in cgw_csum_crc8_rel()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ali Norouzi <[email protected]>

commit b9c310d72783cc2f30d103eed83920a5a29c671a upstream.

cgw_csum_crc8_rel() correctly computes bounds-safe indices via calc_idx():

    int from = calc_idx(crc8->from_idx, cf->len);
    int to   = calc_idx(crc8->to_idx,   cf->len);
    int res  = calc_idx(crc8->result_idx, cf->len);

    if (from < 0 || to < 0 || res < 0)
        return;

However, the loop and the result write then use the raw s8 fields directly instead of the computed variables:

    for (i = crc8->from_idx; ...)        /* BUG: raw negative index */
    cf->data[crc8->result_idx] = ...;    /* BUG: raw negative index */

With from_idx = to_idx = result_idx = -64 on a 64-byte CAN FD frame, calc_idx(-64, 64) = 0 so the guard passes, but the loop iterates with i = -64, reading cf->data[-64], and the write goes to cf->data[-64]. This write might end up to 56 (7.0-rc) or 40 (<= 6.19) bytes before the start of the canfd_frame on the heap.

The companion function cgw_csum_xor_rel() uses from/to/res correctly throughout; fix cgw_csum_crc8_rel() to match.

Confirmed with KASAN on linux-7.0-rc2:

  BUG: KASAN: slab-out-of-bounds in cgw_csum_crc8_rel+0x515/0x5b0
  Read of size 1 at addr ffff8880076619c8 by task poc_cgw_oob/62

To configure the can-gw crc8 checksums CAP_NET_ADMIN is needed.

---
 net/can/gw.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
net/can/gw.c +3 -3
@@ -312,10 +312,10 @@ static void cgw_csum_crc8_rel(struct can
 		return;
 
 	if (from <= to) {
-		for (i = crc8->from_idx; i <= crc8->to_idx; i++)
+		for (i = from; i <= to; i++)
 			crc = crc8->crctab[crc ^ cf->data[i]];
 	} else {
-		for (i = crc8->from_idx; i >= crc8->to_idx; i--)
+		for (i = from; i >= to; i--)
 			crc = crc8->crctab[crc ^ cf->data[i]];
 	}
 
@@ -334,7 +334,7 @@ static void cgw_csum_crc8_rel(struct can
 		break;
 	}
 
-	cf->data[crc8->result_idx] = crc ^ crc8->final_xor_val;
+	cf->data[res] = crc ^ crc8->final_xor_val;
 }
 
 static void cgw_csum_crc8_pos(struct canfd_frame *cf,
author maintainer
#317 lore raw

PATCH 297/491 cpufreq: conservative: Reset requested_freq on limits change

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Viresh Kumar <[email protected]>

commit 6a28fb8cb28b9eb39a392e531d938a889eacafc5 upstream.

A recently reported issue highlighted that the cached requested_freq is not guaranteed to stay in sync with policy->cur. If the platform changes the actual CPU frequency after the governor sets one (e.g. due to platform-specific frequency scaling) and a re-sync occurs later, policy->cur may diverge from requested_freq.

This can lead to incorrect behavior in the conservative governor. For example, the governor may assume the CPU is already running at the maximum frequency and skip further increases even though there is still headroom.

Avoid this by resetting the cached requested_freq to policy->cur on detecting a change in policy limits.

---
 drivers/cpufreq/cpufreq_conservative.c |   12 ++++++++++++
 drivers/cpufreq/cpufreq_governor.c     |    3 +++
 drivers/cpufreq/cpufreq_governor.h     |    1 +
 3 files changed, 16 insertions(+)
drivers/cpufreq/cpufreq_conservative.c +12 -0
@@ -311,6 +311,17 @@ static void cs_start(struct cpufreq_poli
 	dbs_info->requested_freq = policy->cur;
 }
 
+static void cs_limits(struct cpufreq_policy *policy)
+{
+	struct cs_policy_dbs_info *dbs_info = to_dbs_info(policy->governor_data);
+
+	/*
+	 * The limits have changed, so may have the current frequency. Reset
+	 * requested_freq to avoid any unintended outcomes due to the mismatch.
+	 */
+	dbs_info->requested_freq = policy->cur;
+}
+
 static struct dbs_governor cs_governor = {
 	.gov = CPUFREQ_DBS_GOVERNOR_INITIALIZER("conservative"),
 	.kobj_type = { .default_attrs = cs_attributes },
@@ -320,6 +331,7 @@ static struct dbs_governor cs_governor =
 	.init = cs_init,
 	.exit = cs_exit,
 	.start = cs_start,
+	.limits = cs_limits,
 };
 
 #define CPU_FREQ_GOV_CONSERVATIVE	(cs_governor.gov)
drivers/cpufreq/cpufreq_governor.c +3 -0
@@ -555,6 +555,7 @@ EXPORT_SYMBOL_GPL(cpufreq_dbs_governor_s
 
 void cpufreq_dbs_governor_limits(struct cpufreq_policy *policy)
 {
+	struct dbs_governor *gov = dbs_governor_of(policy);
 	struct policy_dbs_info *policy_dbs;
 
 	/* Protect gov->gdbs_data against cpufreq_dbs_governor_exit() */
@@ -566,6 +567,8 @@ void cpufreq_dbs_governor_limits(struct
 	mutex_lock(&policy_dbs->update_mutex);
 	cpufreq_policy_apply_limits(policy);
 	gov_update_sample_delay(policy_dbs, 0);
+	if (gov->limits)
+		gov->limits(policy);
 	mutex_unlock(&policy_dbs->update_mutex);
 
 out:
drivers/cpufreq/cpufreq_governor.h +1 -0
@@ -139,6 +139,7 @@ struct dbs_governor {
 	int (*init)(struct dbs_data *dbs_data);
 	void (*exit)(struct dbs_data *dbs_data);
 	void (*start)(struct cpufreq_policy *policy);
+	void (*limits)(struct cpufreq_policy *policy);
 };
 
 static inline struct dbs_governor *dbs_governor_of(struct cpufreq_policy *policy)
author maintainer
#318 lore raw

PATCH 298/491 media: mc, v4l2: serialize REINIT and REQBUFS with req_queue_mutex

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yuchan Nam <[email protected]>

commit bef4f4a88b73e4cc550d25f665b8a9952af22773 upstream.

MEDIA_REQUEST_IOC_REINIT can run concurrently with VIDIOC_REQBUFS(0) queue teardown paths. This can race request object cleanup against vb2 queue cancellation and lead to use-after-free reports.

We already serialize request queueing against STREAMON/OFF with req_queue_mutex. Extend that serialization to REQBUFS, and also take the same mutex in media_request_ioctl_reinit() so REINIT is in the same exclusion domain.

This keeps request cleanup and queue cancellation from running in parallel for request-capable devices.

Fixes 6093d3002eab ("media: vb2: keep a reference to the request until dqbuf")
---
 drivers/media/mc/mc-request.c        |    5 +++++
 drivers/media/v4l2-core/v4l2-ioctl.c |    5 +++--
 2 files changed, 8 insertions(+), 2 deletions(-)
drivers/media/mc/mc-request.c +5 -0
@@ -190,6 +190,8 @@ static long media_request_ioctl_reinit(s
 	struct media_device *mdev = req->mdev;
 	unsigned long flags;
 
+	mutex_lock(&mdev->req_queue_mutex);
+
 	spin_lock_irqsave(&req->lock, flags);
 	if (req->state != MEDIA_REQUEST_STATE_IDLE &&
 	    req->state != MEDIA_REQUEST_STATE_COMPLETE) {
@@ -197,6 +199,7 @@ static long media_request_ioctl_reinit(s
 			"request: %s not in idle or complete state, cannot reinit\n",
 			req->debug_str);
 		spin_unlock_irqrestore(&req->lock, flags);
+		mutex_unlock(&mdev->req_queue_mutex);
 		return -EBUSY;
 	}
 	if (req->access_count) {
@@ -204,6 +207,7 @@ static long media_request_ioctl_reinit(s
 			"request: %s is being accessed, cannot reinit\n",
 			req->debug_str);
 		spin_unlock_irqrestore(&req->lock, flags);
+		mutex_unlock(&mdev->req_queue_mutex);
 		return -EBUSY;
 	}
 	req->state = MEDIA_REQUEST_STATE_CLEANING;
@@ -214,6 +218,7 @@ static long media_request_ioctl_reinit(s
 	spin_lock_irqsave(&req->lock, flags);
 	req->state = MEDIA_REQUEST_STATE_IDLE;
 	spin_unlock_irqrestore(&req->lock, flags);
+	mutex_unlock(&mdev->req_queue_mutex);
 
 	return 0;
 }
drivers/media/v4l2-core/v4l2-ioctl.c +3 -2
@@ -2963,13 +2963,14 @@ static long __video_do_ioctl(struct file
 		vfh = file->private_data;
 
 	/*
-	 * We need to serialize streamon/off with queueing new requests.
+	 * We need to serialize streamon/off/reqbufs with queueing new requests.
 	 * These ioctls may trigger the cancellation of a streaming
 	 * operation, and that should not be mixed with queueing a new
 	 * request at the same time.
 	 */
 	if (v4l2_device_supports_requests(vfd->v4l2_dev) &&
-	    (cmd == VIDIOC_STREAMON || cmd == VIDIOC_STREAMOFF)) {
+	    (cmd == VIDIOC_STREAMON || cmd == VIDIOC_STREAMOFF ||
+	     cmd == VIDIOC_REQBUFS)) {
 		req_queue_lock = &vfd->v4l2_dev->mdev->req_queue_mutex;
 
 		if (mutex_lock_interruptible(req_queue_lock))
author maintainer
#319 lore raw

PATCH 299/491 virtio_net: Fix UAF on dst_ops when IFF_XMIT_DST_RELEASE is cleared and napi_tx is false

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: xietangxin <[email protected]>

commit ba8bda9a0896746053aa97ac6c3e08168729172c upstream.

A UAF issue occurs when the virtio_net driver is configured with napi_tx=N and the device's IFF_XMIT_DST_RELEASE flag is cleared
(e.g., during the configuration of tc route filter rules).

When IFF_XMIT_DST_RELEASE is removed from the net_device, the network stack expects the driver to hold the reference to skb->dst until the packet is fully transmitted and freed. In virtio_net with napi_tx=N, skbs may remain in the virtio transmit ring for an extended period.

If the network namespace is destroyed while these skbs are still pending, the corresponding dst_ops structure has freed. When a subsequent packet is transmitted, free_old_xmit() is triggered to clean up old skbs. It then calls dst_release() on the skb associated with the stale dst_entry. Since the dst_ops (referenced by the dst_entry) has already been freed, a UAF kernel paging request occurs.

fix it by adds skb_dst_drop(skb) in start_xmit to explicitly release the dst reference before the skb is queued in virtio_net.

Call Trace:
 Unable to handle kernel paging request at virtual address ffff80007e150000
 CPU: 2 UID: 0 PID: 6236 Comm: ping Kdump: loaded Not tainted 7.0.0-rc1+ #6 PREEMPT
  ...
  percpu_counter_add_batch+0x3c/0x158 lib/percpu_counter.c:98 (P)
  dst_release+0xe0/0x110  net/core/dst.c:177
  skb_release_head_state+0xe8/0x108 net/core/skbuff.c:1177
  sk_skb_reason_drop+0x54/0x2d8 net/core/skbuff.c:1255
  dev_kfree_skb_any_reason+0x64/0x78 net/core/dev.c:3469
  napi_consume_skb+0x1c4/0x3a0 net/core/skbuff.c:1527
  __free_old_xmit+0x164/0x230  drivers/net/virtio_net.c:611 [virtio_net]
  free_old_xmit drivers/net/virtio_net.c:1081 [virtio_net]
  start_xmit+0x7c/0x530 drivers/net/virtio_net.c:3329 [virtio_net]
  ...

Reproduction Steps:
NETDEV="enp3s0"

config_qdisc_route_filter() {
tc qdisc del dev $NETDEV root
tc qdisc add dev $NETDEV root handle 1: prio tc filter add dev $NETDEV parent 1:0 \
protocol ip prio 100 route to 100 flowid 1:1
ip route add 192.168.1.100/32 dev $NETDEV realm 100
}

test_ns() {
ip netns add testns
ip link set $NETDEV netns testns
ip netns exec testns ifconfig $NETDEV 10.0.32.46/24 ip netns exec testns ping -c 1 10.0.32.1
ip netns del testns
}

config_qdisc_route_filter

test_ns
sleep 2
test_ns

Fixes f2fc6a54585a ("[NETNS][IPV6] route6 - move ip6_dst_ops inside the network namespace")
Fixes 0287587884b1 ("net: better IFF_XMIT_DST_RELEASE support")
---
 drivers/net/virtio_net.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/virtio_net.c +1 -0
@@ -1740,6 +1740,7 @@ static netdev_tx_t start_xmit(struct sk_
 	/* Don't wait up for transmitted skbs to be freed. */
 	if (!use_napi) {
 		skb_orphan(skb);
+		skb_dst_drop(skb);
 		nf_reset_ct(skb);
 	}
 
author maintainer
#320 lore raw

PATCH 300/491 alarmtimer: Fix argument order in alarm_timer_forward()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhan Xusheng <[email protected]>

commit 5d16467ae56343b9205caedf85e3a131e0914ad8 upstream.

alarm_timer_forward() passes arguments to alarm_forward() in the wrong order:

  alarm_forward(alarm, timr->it_interval, now);

However, alarm_forward() is defined as:

  u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval);

and uses the second argument as the current time:

  delta = ktime_sub(now, alarm->node.expires);

Passing the interval as "now" results in incorrect delta computation, which can lead to missed expirations or incorrect overrun accounting.

This issue has been present since the introduction of alarm_timer_forward().

Fix this by swapping the arguments.

---
 kernel/time/alarmtimer.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/time/alarmtimer.c +1 -1
@@ -606,7 +606,7 @@ static s64 alarm_timer_forward(struct k_
 {
 	struct alarm *alarm = &timr->it.alarm.alarmtimer;
 
-	return alarm_forward(alarm, timr->it_interval, now);
+	return alarm_forward(alarm, now, timr->it_interval);
 }
 
 /**
author maintainer
#321 lore raw

PATCH 301/491 scsi: ibmvfc: Fix OOB access in ibmvfc_discover_targets_done()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tyllis Xu <[email protected]>

commit 61d099ac4a7a8fb11ebdb6e2ec8d77f38e77362f upstream.

A malicious or compromised VIO server can return a num_written value in the discover targets MAD response that exceeds max_targets. This value is stored directly in vhost->num_targets without validation, and is then used as the loop bound in ibmvfc_alloc_targets() to index into disc_buf[], which is only allocated for max_targets entries. Indices at or beyond max_targets access kernel memory outside the DMA-coherent allocation. The out-of-bounds data is subsequently embedded in Implicit Logout and PLOGI MADs that are sent back to the VIO server, leaking kernel memory.

Fix by clamping num_written to max_targets before storing it.

---
 drivers/scsi/ibmvscsi/ibmvfc.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/ibmvscsi/ibmvfc.c +2 -1
@@ -4238,7 +4238,8 @@ static void ibmvfc_discover_targets_done
 	switch (mad_status) {
 	case IBMVFC_MAD_SUCCESS:
 		ibmvfc_dbg(vhost, "Discover Targets succeeded\n");
-		vhost->num_targets = be32_to_cpu(rsp->num_written);
+		vhost->num_targets = min_t(u32, be32_to_cpu(rsp->num_written),
+					   max_targets);
 		ibmvfc_set_host_action(vhost, IBMVFC_HOST_ACTION_ALLOC_TGTS);
 		break;
 	case IBMVFC_MAD_FAILED:
author maintainer
#322 lore raw

PATCH 302/491 scsi: ses: Handle positive SCSI error from ses_recv_diag()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Greg Kroah-Hartman <[email protected]>

commit 7a9f448d44127217fabc4065c5ba070d4e0b5d37 upstream.

ses_recv_diag() can return a positive value, which also means that an error happened, so do not only test for negative values.

---
 drivers/scsi/ses.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/ses.c +1 -1
@@ -184,7 +184,7 @@ static unsigned char *ses_get_page2_desc
 	unsigned char *type_ptr = ses_dev->page1_types;
 	unsigned char *desc_ptr = ses_dev->page2 + 8;
 
-	if (ses_recv_diag(sdev, 2, ses_dev->page2, ses_dev->page2_len) < 0)
+	if (ses_recv_diag(sdev, 2, ses_dev->page2, ses_dev->page2_len))
 		return NULL;
 
 	for (i = 0; i < ses_dev->page1_num_types; i++, type_ptr += 4) {
author maintainer
#323 lore raw

PATCH 303/491 jbd2: gracefully abort on checkpointing state corruptions

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Milos Nikic <[email protected]>

commit bac3190a8e79beff6ed221975e0c9b1b5f2a21da upstream.

This patch targets two internal state machine invariants in checkpoint.c residing inside functions that natively return integer error codes.

- In jbd2_cleanup_journal_tail(): A blocknr of 0 indicates a severely corrupted journal superblock. Replaced the J_ASSERT with a WARN_ON_ONCE and a graceful journal abort, returning -EFSCORRUPTED.

- In jbd2_log_do_checkpoint(): Replaced the J_ASSERT_BH checking for an unexpected buffer_jwrite state. If the warning triggers, we explicitly drop the just-taken get_bh() reference and call __flush_batch() to safely clean up any previously queued buffers in the j_chkpt_bhs array, preventing a memory leak before returning -EFSCORRUPTED.

---
 fs/jbd2/checkpoint.c |   15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)
fs/jbd2/checkpoint.c +13 -2
@@ -303,7 +303,15 @@ restart:
 			 */
 			BUFFER_TRACE(bh, "queue");
 			get_bh(bh);
-			J_ASSERT_BH(bh, !buffer_jwrite(bh));
+			if (WARN_ON_ONCE(buffer_jwrite(bh))) {
+				put_bh(bh); /* drop the ref we just took */
+				spin_unlock(&journal->j_list_lock);
+				/* Clean up any previously batched buffers */
+				if (batch_count)
+					__flush_batch(journal, &batch_count);
+				jbd2_journal_abort(journal, -EFSCORRUPTED);
+				return -EFSCORRUPTED;
+			}
 			journal->j_chkpt_bhs[batch_count++] = bh;
 			transaction->t_chp_stats.cs_written++;
 			transaction->t_checkpoint_list = jh->b_cpnext;
@@ -361,7 +369,10 @@ int jbd2_cleanup_journal_tail(journal_t
 
 	if (!jbd2_journal_get_log_tail(journal, &first_tid, &blocknr))
 		return 1;
-	J_ASSERT(blocknr != 0);
+	if (WARN_ON_ONCE(blocknr == 0)) {
+		jbd2_journal_abort(journal, -EFSCORRUPTED);
+		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * We need to make sure that any blocks that were recently written out
author maintainer
#324 lore raw

PATCH 304/491 ext4: convert inline data to extents when truncate exceeds inline size

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Deepanshu Kartikey <[email protected]>

commit ed9356a30e59c7cc3198e7fc46cfedf3767b9b17 upstream.

Add a check in ext4_setattr() to convert files from inline data storage to extent-based storage when truncate() grows the file size beyond the inline capacity. This prevents the filesystem from entering an inconsistent state where the inline data flag is set but the file size exceeds what can be stored inline.

Without this fix, the following sequence causes a kernel BUG_ON():

1. Mount filesystem with inode that has inline flag set and small size
2. truncate(file, 50MB) - grows size but inline flag remains set
3. sendfile() attempts to write data
4. ext4_write_inline_data() hits BUG_ON(write_size > inline_capacity)

The crash occurs because ext4_write_inline_data() expects inline storage to accommodate the write, but the actual inline capacity (~60 bytes for i_block + ~96 bytes for xattrs) is far smaller than the file size and write request.

The fix checks if the new size from setattr exceeds the inode's actual inline capacity (EXT4_I(inode)->i_inline_size) and converts the file to extent-based storage before proceeding with the size change.

This addresses the root cause by ensuring the inline data flag and file size remain consistent during truncate operations.

---
 fs/ext4/inode.c |   12 ++++++++++++
 1 file changed, 12 insertions(+)
fs/ext4/inode.c +12 -0
@@ -5576,6 +5576,18 @@ int ext4_setattr(struct dentry *dentry,
 		if (IS_I_VERSION(inode) && attr->ia_size != inode->i_size)
 			inode_inc_iversion(inode);
 
+		/*
+		 * If file has inline data but new size exceeds inline capacity,
+		 * convert to extent-based storage first to prevent inconsistent
+		 * state (inline flag set but size exceeds inline capacity).
+		 */
+		if (ext4_has_inline_data(inode) &&
+		    attr->ia_size > EXT4_I(inode)->i_inline_size) {
+			error = ext4_convert_inline_data(inode);
+			if (error)
+				goto err_out;
+		}
+
 		if (shrink) {
 			if (ext4_should_order_data(inode)) {
 				error = ext4_begin_ordered_truncate(inode,
author maintainer
#325 lore raw

PATCH 305/491 ext4: make recently_deleted() properly work with lazy itable initialization

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jan Kara <[email protected]>

commit bd060afa7cc3e0ad30afa9ecc544a78638498555 upstream.

recently_deleted() checks whether inode has been used in the near past. However this can give false positive result when inode table is not initialized yet and we are in fact comparing to random garbage (or stale itable block of a filesystem before mkfs). Ultimately this results in uninitialized inodes being skipped during inode allocation and possibly they are never initialized and thus e2fsck complains. Verify if the inode has been initialized before checking for dtime.

---
 fs/ext4/ialloc.c |    6 ++++++
 1 file changed, 6 insertions(+)
fs/ext4/ialloc.c +6 -0
@@ -686,6 +686,12 @@ static int recently_deleted(struct super
 	if (unlikely(!gdp))
 		return 0;
 
+	/* Inode was never used in this filesystem? */
+	if (ext4_has_group_desc_csum(sb) &&
+	    (gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT) ||
+	     ino >= EXT4_INODES_PER_GROUP(sb) - ext4_itable_unused_count(sb, gdp)))
+		return 0;
+
 	bh = sb_find_get_block(sb, ext4_inode_table(sb, gdp) +
 		       (ino / inodes_per_block));
 	if (!bh || !buffer_uptodate(bh))
author maintainer
#326 lore raw

PATCH 306/491 ext4: avoid allocate block from corrupted group in ext4_mb_find_by_goal()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ye Bin <[email protected]>

commit 46066e3a06647c5b186cc6334409722622d05c44 upstream.

There's issue as follows:
...
EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 206 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 206 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 206 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 206 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 2243 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 2239 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): error count since last fsck: 1
EXT4-fs (mmcblk0p1): initial error at time 1765597433: ext4_mb_generate_buddy:760 EXT4-fs (mmcblk0p1): last error at time 1765597433: ext4_mb_generate_buddy:760
...

According to the log analysis, blocks are always requested from the corrupted block group. This may happen as follows:
ext4_mb_find_by_goal
ext4_mb_load_buddy
ext4_mb_load_buddy_gfp
ext4_mb_init_cache
ext4_read_block_bitmap_nowait
ext4_wait_block_bitmap
ext4_validate_block_bitmap
if (!grp || EXT4_MB_GRP_BBITMAP_CORRUPT(grp))
return -EFSCORRUPTED; // There's no logs.
if (err)
return err; // Will return error
ext4_lock_group(ac->ac_sb, group);
if (unlikely(EXT4_MB_GRP_BBITMAP_CORRUPT(e4b->bd_info))) // Unreachable
goto out;

After commit 9008a58e5dce ("ext4: make the bitmap read routines return real error codes") merged, Commit 163a203ddb36 ("ext4: mark block group as corrupt on block bitmap error") is no real solution for allocating blocks from corrupted block groups. This is because if 'EXT4_MB_GRP_BBITMAP_CORRUPT(e4b->bd_info)' is true, then 'ext4_mb_load_buddy()' may return an error. This means that the block allocation will fail.
Therefore, check block group if corrupted when ext4_mb_load_buddy() returns error.

Fixes 163a203ddb36 ("ext4: mark block group as corrupt on block bitmap error")
Fixes 9008a58e5dce ("ext4: make the bitmap read routines return real error codes")
---
 fs/ext4/mballoc.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
fs/ext4/mballoc.c +5 -1
@@ -1929,8 +1929,12 @@ int ext4_mb_find_by_goal(struct ext4_all
 		return 0;
 
 	err = ext4_mb_load_buddy(ac->ac_sb, group, e4b);
-	if (err)
+	if (err) {
+		if (EXT4_MB_GRP_BBITMAP_CORRUPT(e4b->bd_info) &&
+		    !(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY))
+			return 0;
 		return err;
+	}
 
 	ext4_lock_group(ac->ac_sb, group);
 	if (unlikely(EXT4_MB_GRP_BBITMAP_CORRUPT(e4b->bd_info)))
author maintainer
#327 lore raw

PATCH 307/491 ext4: reject mount if bigalloc with s_first_data_block != 0

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Helen Koike <[email protected]>

commit 3822743dc20386d9897e999dbb990befa3a5b3f8 upstream.

bigalloc with s_first_data_block != 0 is not supported, reject mounting it.

---
 fs/ext4/super.c |    7 +++++++
 1 file changed, 7 insertions(+)
fs/ext4/super.c +7 -0
@@ -3369,6 +3369,13 @@ static int ext4_feature_set_ok(struct su
 			 "extents feature\n");
 		return 0;
 	}
+	if (ext4_has_feature_bigalloc(sb) &&
+	    le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block)) {
+		ext4_msg(sb, KERN_WARNING,
+			 "bad geometry: bigalloc file system with non-zero "
+			 "first_data_block\n");
+		return 0;
+	}
 
 #if !IS_ENABLED(CONFIG_QUOTA) || !IS_ENABLED(CONFIG_QFMT_V2)
 	if (!readonly && (ext4_has_feature_quota(sb) ||
author maintainer
#328 lore raw

PATCH 308/491 phy: ti: j721e-wiz: Fix device node reference leak in wiz_get_lane_phy_types()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Felix Gu <[email protected]>

[ Upstream commit 584b457f4166293bdfa50f930228e9fb91a38392 ]

The serdes device_node is obtained using of_get_child_by_name(), which increments the reference count. However, it is never put, leading to a reference leak.

Add the missing of_node_put() calls to ensure the reference count is properly balanced.

Fixes 7ae14cf581f2 ("phy: ti: j721e-wiz: Implement DisplayPort mode to the wiz driver")
---
 drivers/phy/ti/phy-j721e-wiz.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/phy/ti/phy-j721e-wiz.c +2 -0
index 5536b8f4bfd13..3c0aec368ea9f 100644
@@ -799,6 +799,7 @@ static int wiz_get_lane_phy_types(struct device *dev, struct wiz *wiz)
 			dev_err(dev,
 				"%s: Reading \"reg\" from \"%s\" failed: %d\n",
 				__func__, subnode->name, ret);
+			of_node_put(serdes);
 			return ret;
 		}
 		of_property_read_u32(subnode, "cdns,num-lanes", &num_lanes);
@@ -811,6 +812,7 @@ static int wiz_get_lane_phy_types(struct device *dev, struct wiz *wiz)
 			wiz->lane_phy_type[i] = phy_type;
 	}
 
+	of_node_put(serdes);
 	return 0;
 }
 
Signature · 1 line
2.53.0
author maintainer
#329 lore raw

PATCH 309/491 dmaengine: xilinx: xilinx_dma: Fix dma_device directions

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marek Vasut <[email protected]>

[ Upstream commit e9cc95397bb7da13fe8a5b53a2f23cfaf9018ade ]

Unlike chan->direction , struct dma_device .directions field is a bitfield. Turn chan->direction into a bitfield to make it compatible with struct dma_device .directions .

Fixes 7e01511443c3 ("dmaengine: xilinx_dma: Set dma_device directions")
---
 drivers/dma/xilinx/xilinx_dma.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/dma/xilinx/xilinx_dma.c +1 -1
index 12e9ba5b114db..0d0f3df71245d 100644
@@ -2845,7 +2845,7 @@ static int xilinx_dma_chan_probe(struct xilinx_dma_device *xdev,
 		return -EINVAL;
 	}
 
-	xdev->common.directions |= chan->direction;
+	xdev->common.directions |= BIT(chan->direction);
 
 	/* Request the interrupt */
 	chan->irq = irq_of_parse_and_map(node, chan->tdest);
Signature · 1 line
2.53.0
author maintainer
#330 lore raw

PATCH 310/491 dmaengine: xilinx: xilinx_dma: Fix residue calculation for cyclic DMA

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marek Vasut <[email protected]>

[ Upstream commit f61d145999d61948a23cd436ebbfa4c3b9ab8987 ]

The cyclic DMA calculation is currently entirely broken and reports residue only for the first segment. The problem is twofold.

First, when the first descriptor finishes, it is moved from active_list to done_list, but it is never returned back into the active_list. The xilinx_dma_tx_status() expects the descriptor to be in the active_list to report any meaningful residue information, which never happens after the first descriptor finishes. Fix this up in xilinx_dma_start_transfer() and if the descriptor is cyclic, lift it from done_list and place it back into active_list list.

Second, the segment .status fields of the descriptor remain dirty. Once the DMA did one pass on the descriptor, the .status fields are populated with data by the DMA, but the .status fields are not cleared before reuse during the next cyclic DMA round. The xilinx_dma_get_residue() recognizes that as if the descriptor was complete and had 0 residue, which is bogus. Reinitialize the status field before placing the descriptor back into the active_list.

Fixes c0bba3a99f07 ("dmaengine: vdma: Add Support for Xilinx AXI Direct Memory Access Engine")
---
 drivers/dma/xilinx/xilinx_dma.c | 23 ++++++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)
drivers/dma/xilinx/xilinx_dma.c +22 -1
index 0d0f3df71245d..ca80a1dee8489 100644
@@ -1514,8 +1514,29 @@ static void xilinx_dma_start_transfer(struct xilinx_dma_chan *chan)
 	if (chan->err)
 		return;
 
-	if (list_empty(&chan->pending_list))
+	if (list_empty(&chan->pending_list)) {
+		if (chan->cyclic) {
+			struct xilinx_dma_tx_descriptor *desc;
+			struct list_head *entry;
+
+			desc = list_last_entry(&chan->done_list,
+					       struct xilinx_dma_tx_descriptor, node);
+			list_for_each(entry, &desc->segments) {
+				struct xilinx_axidma_tx_segment *axidma_seg;
+				struct xilinx_axidma_desc_hw *axidma_hw;
+				axidma_seg = list_entry(entry,
+							struct xilinx_axidma_tx_segment,
+							node);
+				axidma_hw = &axidma_seg->hw;
+				axidma_hw->status = 0;
+			}
+
+			list_splice_tail_init(&chan->done_list, &chan->active_list);
+			chan->desc_pendingcount = 0;
+			chan->idle = false;
+		}
 		return;
+	}
 
 	if (!chan->idle)
 		return;
Signature · 1 line
2.53.0
author maintainer
#331 lore raw

PATCH 311/491 dmaengine: xilinx: xilinx_dma: Fix unmasked residue subtraction

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marek Vasut <[email protected]>

[ Upstream commit c7d812e33f3e8ca0fa9eeabf71d1c7bc3acedc09 ]

The segment .control and .status fields both contain top bits which are not part of the buffer size, the buffer size is located only in the bottom max_buffer_len bits. To avoid interference from those top bits, mask out the size using max_buffer_len first, and only then subtract the values.

Fixes a575d0b4e663 ("dmaengine: xilinx_dma: Introduce xilinx_dma_get_residue")
---
 drivers/dma/xilinx/xilinx_dma.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/dma/xilinx/xilinx_dma.c +6 -6
index ca80a1dee8489..a89a150be3284 100644
@@ -964,16 +964,16 @@ static u32 xilinx_dma_get_residue(struct xilinx_dma_chan *chan,
 					      struct xilinx_cdma_tx_segment,
 					      node);
 			cdma_hw = &cdma_seg->hw;
-			residue += (cdma_hw->control - cdma_hw->status) &
-				   chan->xdev->max_buffer_len;
+			residue += (cdma_hw->control & chan->xdev->max_buffer_len) -
+			           (cdma_hw->status & chan->xdev->max_buffer_len);
 		} else if (chan->xdev->dma_config->dmatype ==
 			   XDMA_TYPE_AXIDMA) {
 			axidma_seg = list_entry(entry,
 						struct xilinx_axidma_tx_segment,
 						node);
 			axidma_hw = &axidma_seg->hw;
-			residue += (axidma_hw->control - axidma_hw->status) &
-				   chan->xdev->max_buffer_len;
+			residue += (axidma_hw->control & chan->xdev->max_buffer_len) -
+			           (axidma_hw->status & chan->xdev->max_buffer_len);
 		} else {
 			aximcdma_seg =
 				list_entry(entry,
@@ -981,8 +981,8 @@ static u32 xilinx_dma_get_residue(struct xilinx_dma_chan *chan,
 					   node);
 			aximcdma_hw = &aximcdma_seg->hw;
 			residue +=
-				(aximcdma_hw->control - aximcdma_hw->status) &
-				chan->xdev->max_buffer_len;
+				(aximcdma_hw->control & chan->xdev->max_buffer_len) -
+				(aximcdma_hw->status & chan->xdev->max_buffer_len);
 		}
 	}
 
Signature · 1 line
2.53.0
#332 lore raw

Re: [PATCH 5.10 311/491] dmaengine: xilinx: xilinx_dma: Fix unmasked residue subtraction

Quoted 4 lines · Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

Quoted 8 lines

From: Marek Vasut <[email protected]>

[ Upstream commit c7d812e33f3e8ca0fa9eeabf71d1c7bc3acedc09 ]

The segment .control and .status fields both contain top bits which are not part of the buffer size, the buffer size is located only in the bottom max_buffer_len bits. To avoid interference from those top bits, mask out the size using max_buffer_len first, and only then subtract the values.

This change is harmless, but the problem it claims to fix does not exist.

Ben.

Quoted 45 lines
Fixes a575d0b4e663 ("dmaengine: xilinx_dma: Introduce xilinx_dma_get_residue")
---
 drivers/dma/xilinx/xilinx_dma.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/dma/xilinx/xilinx_dma.c +6 -6
index ca80a1dee8489..a89a150be3284 100644
@@ -964,16 +964,16 @@ static u32 xilinx_dma_get_residue(struct xilinx_dma_chan *chan,
 					      struct xilinx_cdma_tx_segment,
 					      node);
 			cdma_hw = &cdma_seg->hw;
-			residue += (cdma_hw->control - cdma_hw->status) &
-				   chan->xdev->max_buffer_len;
+			residue += (cdma_hw->control & chan->xdev->max_buffer_len) -
+			           (cdma_hw->status & chan->xdev->max_buffer_len);
 		} else if (chan->xdev->dma_config->dmatype ==
 			   XDMA_TYPE_AXIDMA) {
 			axidma_seg = list_entry(entry,
 						struct xilinx_axidma_tx_segment,
 						node);
 			axidma_hw = &axidma_seg->hw;
-			residue += (axidma_hw->control - axidma_hw->status) &
-				   chan->xdev->max_buffer_len;
+			residue += (axidma_hw->control & chan->xdev->max_buffer_len) -
+			           (axidma_hw->status & chan->xdev->max_buffer_len);
 		} else {
 			aximcdma_seg =
 				list_entry(entry,
@@ -981,8 +981,8 @@ static u32 xilinx_dma_get_residue(struct xilinx_dma_chan *chan,
 					   node);
 			aximcdma_hw = &aximcdma_seg->hw;
 			residue +=
-				(aximcdma_hw->control - aximcdma_hw->status) &
-				chan->xdev->max_buffer_len;
+				(aximcdma_hw->control & chan->xdev->max_buffer_len) -
+				(aximcdma_hw->status & chan->xdev->max_buffer_len);
 		}
 	}
 
Signature · 3 lines
Ben Hutchings
It is easier to change the specification to fit the program
than vice versa.
#333 lore raw

Re: [PATCH 5.10 311/491] dmaengine: xilinx: xilinx_dma: Fix unmasked residue subtraction

Quoted 9 lines · Ben Hutchings
Quoted 8 lines · Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marek Vasut <[email protected]>

[ Upstream commit c7d812e33f3e8ca0fa9eeabf71d1c7bc3acedc09 ]

Quoted 7 lines
Quoted 4 lines

The segment .control and .status fields both contain top bits which are not part of the buffer size, the buffer size is located only in the bottom max_buffer_len bits. To avoid interference from those top bits, mask out the size using max_buffer_len first, and only then subtract the values.

This change is harmless, but the problem it claims to fix does not exist.

The current code subtracts two independently read values which both contain status/control MSbits and the actual value LSbits. Depending on the MSbits being identical in both separately read values is unsafe, so the change in this patch masks out the MSbits first and then does the subtraction on the actual value LSbits only, which is safe.

Why do you think the original unsafe behavior can not trigger a failure?

Quoted 47 lines

Ben.

Quoted 45 lines
Fixes a575d0b4e663 ("dmaengine: xilinx_dma: Introduce xilinx_dma_get_residue")
---
  drivers/dma/xilinx/xilinx_dma.c | 12 ++++++------
  1 file changed, 6 insertions(+), 6 deletions(-)
drivers/dma/xilinx/xilinx_dma.c +6 -6
index ca80a1dee8489..a89a150be3284 100644
@@ -964,16 +964,16 @@ static u32 xilinx_dma_get_residue(struct xilinx_dma_chan *chan,
  					      struct xilinx_cdma_tx_segment,
  					      node);
  			cdma_hw = &cdma_seg->hw;
-			residue += (cdma_hw->control - cdma_hw->status) &
-				   chan->xdev->max_buffer_len;
+			residue += (cdma_hw->control & chan->xdev->max_buffer_len) -
+			           (cdma_hw->status & chan->xdev->max_buffer_len);
  		} else if (chan->xdev->dma_config->dmatype ==
  			   XDMA_TYPE_AXIDMA) {
  			axidma_seg = list_entry(entry,
  						struct xilinx_axidma_tx_segment,
  						node);
  			axidma_hw = &axidma_seg->hw;
-			residue += (axidma_hw->control - axidma_hw->status) &
-				   chan->xdev->max_buffer_len;
+			residue += (axidma_hw->control & chan->xdev->max_buffer_len) -
+			           (axidma_hw->status & chan->xdev->max_buffer_len);
  		} else {
  			aximcdma_seg =
  				list_entry(entry,
@@ -981,8 +981,8 @@ static u32 xilinx_dma_get_residue(struct xilinx_dma_chan *chan,
  					   node);
  			aximcdma_hw = &aximcdma_seg->hw;
  			residue +=
-				(aximcdma_hw->control - aximcdma_hw->status) &
-				chan->xdev->max_buffer_len;
+				(aximcdma_hw->control & chan->xdev->max_buffer_len) -
+				(aximcdma_hw->status & chan->xdev->max_buffer_len);
  		}
  	}
 

[...]

#334 lore raw

Re: [PATCH 5.10 311/491] dmaengine: xilinx: xilinx_dma: Fix unmasked residue subtraction

Quoted 18 lines · Marek Vasut
Quoted 9 lines · Ben Hutchings
Quoted 8 lines · Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marek Vasut <[email protected]>

[ Upstream commit c7d812e33f3e8ca0fa9eeabf71d1c7bc3acedc09 ]

Quoted 7 lines
Quoted 4 lines

The segment .control and .status fields both contain top bits which are not part of the buffer size, the buffer size is located only in the bottom max_buffer_len bits. To avoid interference from those top bits, mask out the size using max_buffer_len first, and only then subtract the values.

This change is harmless, but the problem it claims to fix does not exist.

Quoted 7 lines

The current code subtracts two independently read values which both contain status/control MSbits and the actual value LSbits. Depending on the MSbits being identical in both separately read values is unsafe, so the change in this patch masks out the MSbits first and then does the subtraction on the actual value LSbits only, which is safe.

Why do you think the original unsafe behavior can not trigger a failure?

The old code masked out the MSbits after subtraction. So, there was no dependency on their being equal before substraction. Since borrows propagate to the left, not the right, the MSbits could not "interfere" with the LSbits.

If you still aren't convinced, please try to find some example values for which the result would actually change.

Ben.

Signature · 3 lines
Ben Hutchings
It is easier to change the specification to fit the program
than vice versa.
#335 lore raw

Re: [PATCH 5.10 311/491] dmaengine: xilinx: xilinx_dma: Fix unmasked residue subtraction

Quoted 27 lines · Ben Hutchings
Quoted 18 lines · Marek Vasut
Quoted 9 lines · Ben Hutchings
Quoted 8 lines · Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marek Vasut <[email protected]>

[ Upstream commit c7d812e33f3e8ca0fa9eeabf71d1c7bc3acedc09 ]

Quoted 7 lines
Quoted 4 lines

The segment .control and .status fields both contain top bits which are not part of the buffer size, the buffer size is located only in the bottom max_buffer_len bits. To avoid interference from those top bits, mask out the size using max_buffer_len first, and only then subtract the values.

This change is harmless, but the problem it claims to fix does not exist.

Quoted 7 lines

The current code subtracts two independently read values which both contain status/control MSbits and the actual value LSbits. Depending on the MSbits being identical in both separately read values is unsafe, so the change in this patch masks out the MSbits first and then does the subtraction on the actual value LSbits only, which is safe.

Why do you think the original unsafe behavior can not trigger a failure?

Quoted 7 lines

The old code masked out the MSbits after subtraction. So, there was no dependency on their being equal before substraction. Since borrows propagate to the left, not the right, the MSbits could not "interfere" with the LSbits.

If you still aren't convinced, please try to find some example values for which the result would actually change.

Ah sigh, you're right. I will add this into the list of lessons learnt the hard way. Thank you for the clarification.

author maintainer
#336 lore raw

PATCH 312/491 btrfs: fix super block offset in error message in btrfs_validate_super()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mark Harmstone <[email protected]>

[ Upstream commit b52fe51f724385b3ed81e37e510a4a33107e8161 ]

Fix the superblock offset mismatch error message in btrfs_validate_super(): we changed it so that it considers all the superblocks, but the message still assumes we're only looking at the first one.

The change from %u to %llu is because we're changing from a constant to a u64.

Fixes 069ec957c35e ("btrfs: Refactor btrfs_check_super_valid")
---
 fs/btrfs/disk-io.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/btrfs/disk-io.c +2 -2
index 29f0ba4adfbce..7a2b91f6cf14e 100644
@@ -2548,8 +2548,8 @@ static int validate_super(struct btrfs_fs_info *fs_info,
 
 	if (mirror_num >= 0 &&
 	    btrfs_super_bytenr(sb) != btrfs_sb_offset(mirror_num)) {
-		btrfs_err(fs_info, "super offset mismatch %llu != %u",
-			  btrfs_super_bytenr(sb), BTRFS_SUPER_INFO_OFFSET);
+		btrfs_err(fs_info, "super offset mismatch %llu != %llu",
+			  btrfs_super_bytenr(sb), btrfs_sb_offset(mirror_num));
 		ret = -EINVAL;
 	}
 
Signature · 1 line
2.53.0
author maintainer
#337 lore raw

PATCH 313/491 btrfs: fix lost error when running device stats on multiple devices fs

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Filipe Manana <[email protected]>

[ Upstream commit 1c37d896b12dfd0d4c96e310b0033c6676933917 ]

Whenever we get an error updating the device stats item for a device in btrfs_run_dev_stats() we allow the loop to go to the next device, and if updating the stats item for the next device succeeds, we end up losing the error we had from the previous device.

Fix this by breaking out of the loop once we get an error and make sure it's returned to the caller. Since we are in the transaction commit path (and in the critical section actually), returning the error will result in a transaction abort.

Fixes 733f4fbbc108 ("Btrfs: read device stats on mount, write modified ones during commit")
---
 fs/btrfs/volumes.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
fs/btrfs/volumes.c +3 -2
index 2bba6e8d43740..da77493f4c17d 100644
@@ -7553,8 +7553,9 @@ int btrfs_run_dev_stats(struct btrfs_trans_handle *trans)
 		smp_rmb();
 
 		ret = update_dev_stat_item(trans, device);
-		if (!ret)
-			atomic_sub(stats_cnt, &device->dev_stats_ccnt);
+		if (ret)
+			break;
+		atomic_sub(stats_cnt, &device->dev_stats_ccnt);
 	}
 	mutex_unlock(&fs_devices->device_list_mutex);
 
Signature · 1 line
2.53.0
author maintainer
#338 lore raw

PATCH 314/491 dmaengine: xilinx_dma: Program interrupt delay timeout

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Radhey Shyam Pandey <[email protected]>

[ Upstream commit 84b798fedf3fa8f0ab0c096593ba817abc454fe5 ]

Program IRQDelay for AXI DMA. The interrupt timeout mechanism causes the DMA engine to generate an interrupt after the delay time period has expired. It enables dmaengine to respond in real-time even though interrupt coalescing is configured. It also remove the placeholder for delay interrupt and merge it with frame completion interrupt. Since by default interrupt delay timeout is disabled this feature addition has no functional impact on VDMA, MCDMA and CDMA IP's.

Stable-dep-of: a17ce4bc6f4f ("dmaengine: xilinx_dma: Fix reset related timeout with two-channel AXIDMA")

---
 drivers/dma/xilinx/xilinx_dma.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)
drivers/dma/xilinx/xilinx_dma.c +11 -9
index a89a150be3284..7b3c47e442c15 100644
@@ -174,8 +174,10 @@
 #define XILINX_DMA_MAX_TRANS_LEN_MAX	23
 #define XILINX_DMA_V2_MAX_TRANS_LEN_MAX	26
 #define XILINX_DMA_CR_COALESCE_MAX	GENMASK(23, 16)
+#define XILINX_DMA_CR_DELAY_MAX		GENMASK(31, 24)
 #define XILINX_DMA_CR_CYCLIC_BD_EN_MASK	BIT(4)
 #define XILINX_DMA_CR_COALESCE_SHIFT	16
+#define XILINX_DMA_CR_DELAY_SHIFT	24
 #define XILINX_DMA_BD_SOP		BIT(27)
 #define XILINX_DMA_BD_EOP		BIT(26)
 #define XILINX_DMA_COALESCE_MAX		255
@@ -411,6 +413,7 @@ struct xilinx_dma_tx_descriptor {
  * @stop_transfer: Differentiate b/w DMA IP's quiesce
  * @tdest: TDEST value for mcdma
  * @has_vflip: S2MM vertical flip
+ * @irq_delay: Interrupt delay timeout
  */
 struct xilinx_dma_chan {
 	struct xilinx_dma_device *xdev;
@@ -449,6 +452,7 @@ struct xilinx_dma_chan {
 	int (*stop_transfer)(struct xilinx_dma_chan *chan);
 	u16 tdest;
 	bool has_vflip;
+	u8 irq_delay;
 };
 
 /**
@@ -1560,6 +1564,9 @@ static void xilinx_dma_start_transfer(struct xilinx_dma_chan *chan)
 	if (chan->has_sg)
 		xilinx_write(chan, XILINX_DMA_REG_CURDESC,
 			     head_desc->async_tx.phys);
+	reg  &= ~XILINX_DMA_CR_DELAY_MAX;
+	reg  |= chan->irq_delay << XILINX_DMA_CR_DELAY_SHIFT;
+	dma_ctrl_write(chan, XILINX_DMA_REG_DMACR, reg);
 
 	xilinx_dma_start(chan);
 
@@ -1876,15 +1883,8 @@ static irqreturn_t xilinx_dma_irq_handler(int irq, void *data)
 		}
 	}
 
-	if (status & XILINX_DMA_DMASR_DLY_CNT_IRQ) {
-		/*
-		 * Device takes too long to do the transfer when user requires
-		 * responsiveness.
-		 */
-		dev_dbg(chan->dev, "Inter-packet latency too long\n");
-	}
-
-	if (status & XILINX_DMA_DMASR_FRM_CNT_IRQ) {
+	if (status & (XILINX_DMA_DMASR_FRM_CNT_IRQ |
+		      XILINX_DMA_DMASR_DLY_CNT_IRQ)) {
 		spin_lock(&chan->lock);
 		xilinx_dma_complete_descriptor(chan);
 		chan->idle = true;
@@ -2801,6 +2801,8 @@ static int xilinx_dma_chan_probe(struct xilinx_dma_device *xdev,
 	/* Retrieve the channel properties from the device tree */
 	has_dre = of_property_read_bool(node, "xlnx,include-dre");
 
+	of_property_read_u8(node, "xlnx,irq-delay", &chan->irq_delay);
+
 	chan->genlock = of_property_read_bool(node, "xlnx,genlock-mode");
 
 	err = of_property_read_u32(node, "xlnx,datawidth", &value);
Signature · 1 line
2.53.0
author maintainer
#339 lore raw

PATCH 315/491 dmaengine: xilinx_dma: Fix reset related timeout with two-channel AXIDMA

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tomi Valkeinen <[email protected]>

[ Upstream commit a17ce4bc6f4f9acf77ba416c36791a15602e53aa ]

A single AXIDMA controller can have one or two channels. When it has two channels, the reset for both are tied together: resetting one channel resets the other as well. This creates a problem where resetting one channel will reset the registers for both channels, including clearing interrupt enable bits for the other channel, which can then lead to timeouts as the driver is waiting for an interrupt which never comes.

The driver currently has a probe-time work around for this: when a channel is created, the driver also resets and enables the interrupts. With two channels the reset for the second channel will clear the interrupt enables for the first one. The work around in the driver is just to manually enable the interrupts again in xilinx_dma_alloc_chan_resources().

This workaround only addresses the probe-time issue. When channels are reset at runtime (e.g., in xilinx_dma_terminate_all() or during error recovery), there's no corresponding mechanism to restore the other channel's interrupt enables. This leads to one channel having its interrupts disabled while the driver expects them to work, causing timeouts and DMA failures.

A proper fix is a complicated matter, as we should not reset the other channel when it's operating normally. So, perhaps, there should be some kind of synchronization for a common reset, which is not trivial to implement. To add to the complexity, the driver also supports other DMA types, like VDMA, CDMA and MCDMA, which don't have a shared reset.

However, when the two-channel AXIDMA is used in the (assumably) normal use case, providing DMA for a single memory-to-memory device, the common reset is a bit smaller issue: when something bad happens on one channel, or when one channel is terminated, the assumption is that we also want to terminate the other channel. And thus resetting both at the same time is "ok".

With that line of thinking we can implement a bit better work around than just the current probe time work around: let's enable the AXIDMA interrupts at xilinx_dma_start_transfer() instead. This ensures interrupts are enabled whenever a transfer starts, regardless of any prior resets that may have cleared them.

This approach is also more logical: enable interrupts only when needed for a transfer, rather than at resource allocation time, and, I think, all the other DMA types should also use this model, but I'm reluctant to do such changes as I cannot test them.

The reset function still enables interrupts even though it's not needed for AXIDMA anymore, but it's common code for all DMA types (VDMA, CDMA, MCDMA), so leave it unchanged to avoid affecting other variants.

Fixes c0bba3a99f07 ("dmaengine: vdma: Add Support for Xilinx AXI Direct Memory Access Engine")
---
 drivers/dma/xilinx/xilinx_dma.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)
drivers/dma/xilinx/xilinx_dma.c +1 -8
index 7b3c47e442c15..ff7fda42b5ca2 100644
@@ -1190,14 +1190,6 @@ static int xilinx_dma_alloc_chan_resources(struct dma_chan *dchan)
 
 	dma_cookie_init(dchan);
 
-	if (chan->xdev->dma_config->dmatype == XDMA_TYPE_AXIDMA) {
-		/* For AXI DMA resetting once channel will reset the
-		 * other channel as well so enable the interrupts here.
-		 */
-		dma_ctrl_set(chan, XILINX_DMA_REG_DMACR,
-			      XILINX_DMA_DMAXR_ALL_IRQ_MASK);
-	}
-
 	if ((chan->xdev->dma_config->dmatype == XDMA_TYPE_CDMA) && chan->has_sg)
 		dma_ctrl_set(chan, XILINX_DMA_REG_DMACR,
 			     XILINX_CDMA_CR_SGMODE);
@@ -1566,6 +1558,7 @@ static void xilinx_dma_start_transfer(struct xilinx_dma_chan *chan)
 			     head_desc->async_tx.phys);
 	reg  &= ~XILINX_DMA_CR_DELAY_MAX;
 	reg  |= chan->irq_delay << XILINX_DMA_CR_DELAY_SHIFT;
+	reg |= XILINX_DMA_DMAXR_ALL_IRQ_MASK;
 	dma_ctrl_write(chan, XILINX_DMA_REG_DMACR, reg);
 
 	xilinx_dma_start(chan);
Signature · 1 line
2.53.0
author maintainer
#340 lore raw

PATCH 316/491 futex: Clear stale exiting pointer in futex_lock_pi() retry path

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Davidlohr Bueso <[email protected]>

commit 210d36d892de5195e6766c45519dfb1e65f3eb83 upstream.

Fuzzying/stressing futexes triggered:

    WARNING: kernel/futex/core.c:825 at wait_for_owner_exiting+0x7a/0x80, CPU#11: futex_lock_pi_s/524

When futex_lock_pi_atomic() sees the owner is exiting, it returns -EBUSY and stores a refcounted task pointer in 'exiting'.

After wait_for_owner_exiting() consumes that reference, the local pointer is never reset to nil. Upon a retry, if futex_lock_pi_atomic() returns a different error, the bogus pointer is passed to wait_for_owner_exiting().

CPU0 CPU1 CPU2
futex_lock_pi(uaddr)
// acquires the PI futex
exit()

    futex_cleanup_begin()
      futex_state = EXITING;
			     futex_lock_pi(uaddr)
			       futex_lock_pi_atomic()
				 attach_to_pi_owner()
				   // observes EXITING
				   *exiting = owner;  // takes ref
				   return -EBUSY
			       wait_for_owner_exiting(-EBUSY, owner)
				 put_task_struct();   // drops ref
			       // exiting still points to owner
			       goto retry;
			       futex_lock_pi_atomic()
				 lock_pi_update_atomic()
				   cmpxchg(uaddr)
					*uaddr ^= WAITERS // whatever
				   // value changed
				 return -EAGAIN;
			       wait_for_owner_exiting(-EAGAIN, exiting) // stale
				 WARN_ON_ONCE(exiting)

Fix this by resetting upon retry, essentially aligning it with requeue_pi.

---
 kernel/futex/core.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
kernel/futex/core.c +2 -1
@@ -2785,9 +2785,9 @@ static int futex_lock_pi(u32 __user *uad
 			 ktime_t *time, int trylock)
 {
 	struct hrtimer_sleeper timeout, *to;
-	struct task_struct *exiting = NULL;
 	struct rt_mutex_waiter rt_waiter;
 	struct futex_hash_bucket *hb;
+	struct task_struct *exiting;
 	struct futex_q q = futex_q_init;
 	int res, ret;
 
@@ -2800,6 +2800,7 @@ static int futex_lock_pi(u32 __user *uad
 	to = futex_setup_timer(time, &timeout, FLAGS_CLOCKRT, 0);
 
 retry:
+	exiting = NULL;
 	ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q.key, FUTEX_WRITE);
 	if (unlikely(ret != 0))
 		goto out;
author maintainer
#341 lore raw

PATCH 317/491 HID: wacom: fix out-of-bounds read in wacom_intuos_bt_irq

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Benoît Sevens <[email protected]>

[ Upstream commit 2f1763f62909ccb6386ac50350fa0abbf5bb16a9 ]

The wacom_intuos_bt_irq() function processes Bluetooth HID reports without sufficient bounds checking. A maliciously crafted short report can trigger an out-of-bounds read when copying data into the wacom structure.

Specifically, report 0x03 requires at least 22 bytes to safely read the processed data and battery status, while report 0x04 (which falls through to 0x03) requires 32 bytes.

Add explicit length checks for these report IDs and log a warning if a short report is received.

---
 drivers/hid/wacom_wac.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
drivers/hid/wacom_wac.c +10 -0
index 2a7cd5be87444..12d70983ed407 100644
@@ -1251,10 +1251,20 @@ static int wacom_intuos_bt_irq(struct wacom_wac *wacom, size_t len)
 
 	switch (data[0]) {
 	case 0x04:
+		if (len < 32) {
+			dev_warn(wacom->pen_input->dev.parent,
+				 "Report 0x04 too short: %zu bytes\n", len);
+			break;
+		}
 		wacom_intuos_bt_process_data(wacom, data + i);
 		i += 10;
 		fallthrough;
 	case 0x03:
+		if (i == 1 && len < 22) {
+			dev_warn(wacom->pen_input->dev.parent,
+				 "Report 0x03 too short: %zu bytes\n", len);
+			break;
+		}
 		wacom_intuos_bt_process_data(wacom, data + i);
 		i += 10;
 		wacom_intuos_bt_process_data(wacom, data + i);
Signature · 1 line
2.53.0
author maintainer
#342 lore raw

PATCH 318/491 atm: lec: fix use-after-free in sock_def_readable()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Deepanshu Kartikey <[email protected]>

[ Upstream commit 922814879542c2e397b0e9641fd36b8202a8e555 ]

A race condition exists between lec_atm_close() setting priv->lecd to NULL and concurrent access to priv->lecd in send_to_lecd(), lec_handle_bridge(), and lec_atm_send(). When the socket is freed via RCU while another thread is still using it, a use-after-free occurs in sock_def_readable() when accessing the socket's wait queue.

The root cause is that lec_atm_close() clears priv->lecd without any synchronization, while callers dereference priv->lecd without any protection against concurrent teardown.

Fix this by converting priv->lecd to an RCU-protected pointer:
- Mark priv->lecd as __rcu in lec.h
- Use rcu_assign_pointer() in lec_atm_close() and lecd_attach() for safe pointer assignment
- Use rcu_access_pointer() for NULL checks that do not dereference the pointer in lec_start_xmit(), lec_push(), send_to_lecd() and lecd_attach()
- Use rcu_read_lock/rcu_dereference/rcu_read_unlock in send_to_lecd(), lec_handle_bridge() and lec_atm_send() to safely access lecd
- Use rcu_assign_pointer() followed by synchronize_rcu() in lec_atm_close() to ensure all readers have completed before proceeding. This is safe since lec_atm_close() is called from vcc_release() which holds lock_sock(), a sleeping lock.
- Remove the manual sk_receive_queue drain from lec_atm_close() since vcc_destroy_socket() already drains it after lec_atm_close() returns.

v2: Switch from spinlock + sock_hold/put approach to RCU to properly
fix the race. The v1 spinlock approach had two issues pointed out by Eric Dumazet:
1. priv->lecd was still accessed directly after releasing the lock instead of using a local copy.
2. The spinlock did not prevent packets being queued after lec_atm_close() drains sk_receive_queue since timer and workqueue paths bypass netif_stop_queue().

Note: Syzbot patch testing was attempted but the test VM terminated
unexpectedly with "Connection to localhost closed by remote host", likely due to a QEMU AHCI emulation issue unrelated to this fix. Compile testing with "make W=1 net/atm/lec.o" passes cleanly.

---
 net/atm/lec.c |   72 +++++++++++++++++++++++++++++++++++++---------------------
 net/atm/lec.h |    2 -
 2 files changed, 48 insertions(+), 26 deletions(-)
net/atm/lec.c +47 -25
@@ -142,6 +142,7 @@ static void lec_handle_bridge(struct sk_
 		struct sock *sk;
 		struct sk_buff *skb2;
 		struct atmlec_msg *mesg;
+		struct atm_vcc *vcc;
 
 		skb2 = alloc_skb(sizeof(struct atmlec_msg), GFP_ATOMIC);
 		if (skb2 == NULL)
@@ -154,10 +155,18 @@ static void lec_handle_bridge(struct sk_
 					/* 0x01 is topology change */
 
 		priv = netdev_priv(dev);
-		atm_force_charge(priv->lecd, skb2->truesize);
-		sk = sk_atm(priv->lecd);
-		skb_queue_tail(&sk->sk_receive_queue, skb2);
-		sk->sk_data_ready(sk);
+
+		rcu_read_lock();
+		vcc = rcu_dereference(priv->lecd);
+		if (vcc) {
+			atm_force_charge(vcc, skb2->truesize);
+			sk = sk_atm(vcc);
+			skb_queue_tail(&sk->sk_receive_queue, skb2);
+			sk->sk_data_ready(sk);
+		} else {
+			dev_kfree_skb(skb2);
+		}
+		rcu_read_unlock();
 	}
 }
 #endif /* IS_ENABLED(CONFIG_BRIDGE) */
@@ -216,7 +225,7 @@ static netdev_tx_t lec_start_xmit(struct
 	int is_rdesc;
 
 	pr_debug("called\n");
-	if (!priv->lecd) {
+	if (!rcu_access_pointer(priv->lecd)) {
 		pr_info("%s:No lecd attached\n", dev->name);
 		dev->stats.tx_errors++;
 		netif_stop_queue(dev);
@@ -443,6 +452,7 @@ static int lec_atm_send(struct atm_vcc *
 			/* hit from bridge table, send LE_ARP_RESPONSE */
 			struct sk_buff *skb2;
 			struct sock *sk;
+			struct atm_vcc *vcc;
 
 			pr_debug("%s: entry found, responding to zeppelin\n",
 				 dev->name);
@@ -451,10 +461,18 @@ static int lec_atm_send(struct atm_vcc *
 				break;
 			skb2->len = sizeof(struct atmlec_msg);
 			skb_copy_to_linear_data(skb2, mesg, sizeof(*mesg));
-			atm_force_charge(priv->lecd, skb2->truesize);
-			sk = sk_atm(priv->lecd);
-			skb_queue_tail(&sk->sk_receive_queue, skb2);
-			sk->sk_data_ready(sk);
+
+			rcu_read_lock();
+			vcc = rcu_dereference(priv->lecd);
+			if (vcc) {
+				atm_force_charge(vcc, skb2->truesize);
+				sk = sk_atm(vcc);
+				skb_queue_tail(&sk->sk_receive_queue, skb2);
+				sk->sk_data_ready(sk);
+			} else {
+				dev_kfree_skb(skb2);
+			}
+			rcu_read_unlock();
 		}
 	}
 #endif /* IS_ENABLED(CONFIG_BRIDGE) */
@@ -470,23 +488,16 @@ static int lec_atm_send(struct atm_vcc *
 
 static void lec_atm_close(struct atm_vcc *vcc)
 {
-	struct sk_buff *skb;
 	struct net_device *dev = (struct net_device *)vcc->proto_data;
 	struct lec_priv *priv = netdev_priv(dev);
 
-	priv->lecd = NULL;
+	rcu_assign_pointer(priv->lecd, NULL);
+	synchronize_rcu();
 	/* Do something needful? */
 
 	netif_stop_queue(dev);
 	lec_arp_destroy(priv);
 
-	if (skb_peek(&sk_atm(vcc)->sk_receive_queue))
-		pr_info("%s closing with messages pending\n", dev->name);
-	while ((skb = skb_dequeue(&sk_atm(vcc)->sk_receive_queue))) {
-		atm_return(vcc, skb->truesize);
-		dev_kfree_skb(skb);
-	}
-
 	pr_info("%s: Shut down!\n", dev->name);
 	module_put(THIS_MODULE);
 }
@@ -512,12 +523,14 @@ send_to_lecd(struct lec_priv *priv, atml
 	     const unsigned char *mac_addr, const unsigned char *atm_addr,
 	     struct sk_buff *data)
 {
+	struct atm_vcc *vcc;
 	struct sock *sk;
 	struct sk_buff *skb;
 	struct atmlec_msg *mesg;
 
-	if (!priv || !priv->lecd)
+	if (!priv || !rcu_access_pointer(priv->lecd))
 		return -1;
+
 	skb = alloc_skb(sizeof(struct atmlec_msg), GFP_ATOMIC);
 	if (!skb)
 		return -1;
@@ -534,18 +547,27 @@ send_to_lecd(struct lec_priv *priv, atml
 	if (atm_addr)
 		memcpy(&mesg->content.normal.atm_addr, atm_addr, ATM_ESA_LEN);
 
-	atm_force_charge(priv->lecd, skb->truesize);
-	sk = sk_atm(priv->lecd);
+	rcu_read_lock();
+	vcc = rcu_dereference(priv->lecd);
+	if (!vcc) {
+		rcu_read_unlock();
+		kfree_skb(skb);
+		return -1;
+	}
+
+	atm_force_charge(vcc, skb->truesize);
+	sk = sk_atm(vcc);
 	skb_queue_tail(&sk->sk_receive_queue, skb);
 	sk->sk_data_ready(sk);
 
 	if (data != NULL) {
 		pr_debug("about to send %d bytes of data\n", data->len);
-		atm_force_charge(priv->lecd, data->truesize);
+		atm_force_charge(vcc, data->truesize);
 		skb_queue_tail(&sk->sk_receive_queue, data);
 		sk->sk_data_ready(sk);
 	}
 
+	rcu_read_unlock();
 	return 0;
 }
 
@@ -620,7 +642,7 @@ static void lec_push(struct atm_vcc *vcc
 
 		atm_return(vcc, skb->truesize);
 		if (*(__be16 *) skb->data == htons(priv->lecid) ||
-		    !priv->lecd || !(dev->flags & IFF_UP)) {
+		    !rcu_access_pointer(priv->lecd) || !(dev->flags & IFF_UP)) {
 			/*
 			 * Probably looping back, or if lecd is missing,
 			 * lecd has gone down
@@ -755,12 +777,12 @@ static int lecd_attach(struct atm_vcc *v
 		priv = netdev_priv(dev_lec[i]);
 	} else {
 		priv = netdev_priv(dev_lec[i]);
-		if (priv->lecd)
+		if (rcu_access_pointer(priv->lecd))
 			return -EADDRINUSE;
 	}
 	lec_arp_init(priv);
 	priv->itfnum = i;	/* LANE2 addition */
-	priv->lecd = vcc;
+	rcu_assign_pointer(priv->lecd, vcc);
 	vcc->dev = &lecatm_dev;
 	vcc_insert_socket(sk_atm(vcc));
 
net/atm/lec.h +1 -1
@@ -91,7 +91,7 @@ struct lec_priv {
 						 */
 	spinlock_t lec_arp_lock;
 	struct atm_vcc *mcast_vcc;		/* Default Multicast Send VCC */
-	struct atm_vcc *lecd;
+	struct atm_vcc __rcu *lecd;
 	struct delayed_work lec_arp_work;	/* C10 */
 	unsigned int maximum_unknown_frame_count;
 						/*
author maintainer
#343 lore raw

PATCH 319/491 objtool: Fix Clang jump table detection

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Josh Poimboeuf <[email protected]>

[ Upstream commit 4e5019216402ad0b4a84cff457b662d26803f103 ]

With Clang, there can be a conditional forward jump between the load of the jump table address and the indirect branch.

Fixes the following warning:

vmlinux.o: warning: objtool: ___bpf_prog_run+0x1c5: sibling call from callable instruction with modified stack frame

---
 tools/objtool/check.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
tools/objtool/check.c +2 -3
index 20ccdd60353be..ac4bd2d4fdda6 100644
@@ -1664,12 +1664,11 @@ static void mark_func_jump_tables(struct objtool_file *file,
 			last = insn;
 
 		/*
-		 * Store back-pointers for unconditional forward jumps such
+		 * Store back-pointers for forward jumps such
 		 * that find_jump_table() can back-track using those and
 		 * avoid some potentially confusing code.
 		 */
-		if (insn->type == INSN_JUMP_UNCONDITIONAL && insn->jump_dest &&
-		    insn->offset > last->offset &&
+		if (insn->jump_dest &&
 		    insn->jump_dest->offset > insn->offset &&
 		    !insn->jump_dest->first_jump_src) {
 
Signature · 1 line
2.53.0
author maintainer
#344 lore raw

PATCH 320/491 HID: multitouch: Check to ensure report responses match the request

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Lee Jones <[email protected]>

[ Upstream commit e716edafedad4952fe3a4a273d2e039a84e8681a ]

It is possible for a malicious (or clumsy) device to respond to a specific report's feature request using a completely different report ID. This can cause confusion in the HID core resulting in nasty side-effects such as OOB writes.

Add a check to ensure that the report ID in the response, matches the one that was requested. If it doesn't, omit reporting the raw event and return early.

---
 drivers/hid/hid-multitouch.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/hid/hid-multitouch.c +7 -0
index 227cf3f6ca227..948bd59ab5d21 100644
@@ -442,12 +442,19 @@ static void mt_get_feature(struct hid_device *hdev, struct hid_report *report)
 		dev_warn(&hdev->dev, "failed to fetch feature %d\n",
 			 report->id);
 	} else {
+		/* The report ID in the request and the response should match */
+		if (report->id != buf[0]) {
+			hid_err(hdev, "Returned feature report did not match the request\n");
+			goto free;
+		}
+
 		ret = hid_report_raw_event(hdev, HID_FEATURE_REPORT, buf,
 					   size, 0);
 		if (ret)
 			dev_warn(&hdev->dev, "failed to report feature\n");
 	}
 
+free:
 	kfree(buf);
 }
 
Signature · 1 line
2.53.0
author maintainer
#345 lore raw

PATCH 321/491 crypto: af-alg - fix NULL pointer dereference in scatterwalk

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Norbert Szetei <[email protected]>

[ Upstream commit 62397b493e14107ae82d8b80938f293d95425bcb ]

The AF_ALG interface fails to unmark the end of a Scatter/Gather List (SGL) when chaining a new af_alg_tsgl structure. If a sendmsg() fills an SGL exactly to MAX_SGL_ENTS, the last entry is marked as the end. A subsequent sendmsg() allocates a new SGL and chains it, but fails to clear the end marker on the previous SGL's last data entry.

This causes the crypto scatterwalk to hit a premature end, returning NULL on sg_next() and leading to a kernel panic during dereference.

Fix this by explicitly unmarking the end of the previous SGL when performing sg_chain() in af_alg_alloc_tsgl().

Fixes 8ff590903d5f ("crypto: algif_skcipher - User-space interface for skcipher operations")
---
 crypto/af_alg.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
crypto/af_alg.c +3 -1
index 3d622904f4c3d..5e0fe58f8bdd9 100644
@@ -515,8 +515,10 @@ static int af_alg_alloc_tsgl(struct sock *sk)
 		sg_init_table(sgl->sg, MAX_SGL_ENTS + 1);
 		sgl->cur = 0;
 
-		if (sg)
+		if (sg) {
+			sg_unmark_end(sg + MAX_SGL_ENTS - 1);
 			sg_chain(sg, MAX_SGL_ENTS + 1, sgl->sg);
+		}
 
 		list_add_tail(&sgl->list, &ctx->tsgl_list);
 	}
Signature · 1 line
2.53.0
author maintainer
#346 lore raw

PATCH 322/491 net: qrtr: Add GFP flags parameter to qrtr_alloc_ctrl_packet

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Loic Poulain <[email protected]>

[ Upstream commit f7dec6cb914c8923808eefc034bba8227a5dfe3a ]

This will be requested for allocating control packet in atomic context.

Stable-dep-of: 2428083101f6 ("net: qrtr: replace qrtr_tx_flow radix_tree with xarray to fix memory leak")

---
 net/qrtr/af_qrtr.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
net/qrtr/af_qrtr.c +7 -5
index 8476a229bce0a..01d8cb93ad756 100644
@@ -540,18 +540,20 @@ EXPORT_SYMBOL_GPL(qrtr_endpoint_post);
 /**
  * qrtr_alloc_ctrl_packet() - allocate control packet skb
  * @pkt: reference to qrtr_ctrl_pkt pointer
+ * @flags: the type of memory to allocate
  *
  * Returns newly allocated sk_buff, or NULL on failure
  *
  * This function allocates a sk_buff large enough to carry a qrtr_ctrl_pkt and
  * on success returns a reference to the control packet in @pkt.
  */
-static struct sk_buff *qrtr_alloc_ctrl_packet(struct qrtr_ctrl_pkt **pkt)
+static struct sk_buff *qrtr_alloc_ctrl_packet(struct qrtr_ctrl_pkt **pkt,
+					      gfp_t flags)
 {
 	const int pkt_len = sizeof(struct qrtr_ctrl_pkt);
 	struct sk_buff *skb;
 
-	skb = alloc_skb(QRTR_HDR_MAX_SIZE + pkt_len, GFP_KERNEL);
+	skb = alloc_skb(QRTR_HDR_MAX_SIZE + pkt_len, flags);
 	if (!skb)
 		return NULL;
 
@@ -620,7 +622,7 @@ void qrtr_endpoint_unregister(struct qrtr_endpoint *ep)
 	mutex_unlock(&node->ep_lock);
 
 	/* Notify the local controller about the event */
-	skb = qrtr_alloc_ctrl_packet(&pkt);
+	skb = qrtr_alloc_ctrl_packet(&pkt, GFP_KERNEL);
 	if (skb) {
 		pkt->cmd = cpu_to_le32(QRTR_TYPE_BYE);
 		qrtr_local_enqueue(NULL, skb, QRTR_TYPE_BYE, &src, &dst);
@@ -677,7 +679,7 @@ static void qrtr_port_remove(struct qrtr_sock *ipc)
 	to.sq_node = QRTR_NODE_BCAST;
 	to.sq_port = QRTR_PORT_CTRL;
 
-	skb = qrtr_alloc_ctrl_packet(&pkt);
+	skb = qrtr_alloc_ctrl_packet(&pkt, GFP_KERNEL);
 	if (skb) {
 		pkt->cmd = cpu_to_le32(QRTR_TYPE_DEL_CLIENT);
 		pkt->client.node = cpu_to_le32(ipc->us.sq_node);
@@ -992,7 +994,7 @@ static int qrtr_send_resume_tx(struct qrtr_cb *cb)
 	if (!node)
 		return -EINVAL;
 
-	skb = qrtr_alloc_ctrl_packet(&pkt);
+	skb = qrtr_alloc_ctrl_packet(&pkt, GFP_KERNEL);
 	if (!skb)
 		return -ENOMEM;
 
Signature · 1 line
2.53.0
author maintainer
#347 lore raw

PATCH 323/491 net: qrtr: Release distant nodes along the bridge node

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Loic Poulain <[email protected]>

[ Upstream commit 90829f07baea7efe4ca77f79820f3849081186a8 ]

Distant QRTR nodes can be accessed via an other node that acts as a bridge. When the a QRTR endpoint associated to a bridge node is released, all the linked distant nodes should also be released.

This patch fixes endpoint release by:
- Submitting QRTR BYE message locally on behalf of all the nodes accessible through the endpoint.
- Removing all the routable node IDs from radix tree pointing to the released node endpoint.

Stable-dep-of: 2428083101f6 ("net: qrtr: replace qrtr_tx_flow radix_tree with xarray to fix memory leak")

---
 net/qrtr/af_qrtr.c | 25 +++++++++++++++++++------
 1 file changed, 19 insertions(+), 6 deletions(-)
net/qrtr/af_qrtr.c +19 -6
index 01d8cb93ad756..68b6124fc8a10 100644
@@ -172,8 +172,13 @@ static void __qrtr_node_release(struct kref *kref)
 	void __rcu **slot;
 
 	spin_lock_irqsave(&qrtr_nodes_lock, flags);
-	if (node->nid != QRTR_EP_NID_AUTO)
-		radix_tree_delete(&qrtr_nodes, node->nid);
+	/* If the node is a bridge for other nodes, there are possibly
+	 * multiple entries pointing to our released node, delete them all.
+	 */
+	radix_tree_for_each_slot(slot, &qrtr_nodes, &iter, 0) {
+		if (*slot == node)
+			radix_tree_iter_delete(&qrtr_nodes, &iter, slot);
+	}
 	spin_unlock_irqrestore(&qrtr_nodes_lock, flags);
 
 	list_del(&node->item);
@@ -615,6 +620,7 @@ void qrtr_endpoint_unregister(struct qrtr_endpoint *ep)
 	struct qrtr_ctrl_pkt *pkt;
 	struct qrtr_tx_flow *flow;
 	struct sk_buff *skb;
+	unsigned long flags;
 	void __rcu **slot;
 
 	mutex_lock(&node->ep_lock);
@@ -622,11 +628,18 @@ void qrtr_endpoint_unregister(struct qrtr_endpoint *ep)
 	mutex_unlock(&node->ep_lock);
 
 	/* Notify the local controller about the event */
-	skb = qrtr_alloc_ctrl_packet(&pkt, GFP_KERNEL);
-	if (skb) {
-		pkt->cmd = cpu_to_le32(QRTR_TYPE_BYE);
-		qrtr_local_enqueue(NULL, skb, QRTR_TYPE_BYE, &src, &dst);
+	spin_lock_irqsave(&qrtr_nodes_lock, flags);
+	radix_tree_for_each_slot(slot, &qrtr_nodes, &iter, 0) {
+		if (*slot != node)
+			continue;
+		src.sq_node = iter.index;
+		skb = qrtr_alloc_ctrl_packet(&pkt, GFP_ATOMIC);
+		if (skb) {
+			pkt->cmd = cpu_to_le32(QRTR_TYPE_BYE);
+			qrtr_local_enqueue(NULL, skb, QRTR_TYPE_BYE, &src, &dst);
+		}
 	}
+	spin_unlock_irqrestore(&qrtr_nodes_lock, flags);
 
 	/* Wake up any transmitters waiting for resume-tx from the node */
 	mutex_lock(&node->qrtr_tx_lock);
Signature · 1 line
2.53.0
author maintainer
#348 lore raw

PATCH 324/491 net: qrtr: replace qrtr_tx_flow radix_tree with xarray to fix memory leak

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jiayuan Chen <[email protected]>

[ Upstream commit 2428083101f6883f979cceffa76cd8440751ffe6 ]

__radix_tree_create() allocates and links intermediate nodes into the tree one by one. If a subsequent allocation fails, the already-linked nodes remain in the tree with no corresponding leaf entry. These orphaned internal nodes are never reclaimed because radix_tree_for_each_slot() only visits slots containing leaf values.

The radix_tree API is deprecated in favor of xarray. As suggested by Matthew Wilcox, migrate qrtr_tx_flow from radix_tree to xarray instead of fixing the radix_tree itself [1]. xarray properly handles cleanup of internal nodes — xa_destroy() frees all internal xarray nodes when the qrtr_node is released, preventing the leak.

[1] https://lore.kernel.org/all/[email protected]/T/

---
 net/qrtr/af_qrtr.c | 31 +++++++++++++------------------
 1 file changed, 13 insertions(+), 18 deletions(-)
net/qrtr/af_qrtr.c +13 -18
index 68b6124fc8a10..fdb7a5a12f035 100644
@@ -116,7 +116,7 @@ static DEFINE_XARRAY_ALLOC(qrtr_ports);
  * @ep: endpoint
  * @ref: reference count for node
  * @nid: node id
- * @qrtr_tx_flow: tree of qrtr_tx_flow, keyed by node << 32 | port
+ * @qrtr_tx_flow: xarray of qrtr_tx_flow, keyed by node << 32 | port
  * @qrtr_tx_lock: lock for qrtr_tx_flow inserts
  * @rx_queue: receive queue
  * @item: list item for broadcast list
@@ -127,7 +127,7 @@ struct qrtr_node {
 	struct kref ref;
 	unsigned int nid;
 
-	struct radix_tree_root qrtr_tx_flow;
+	struct xarray qrtr_tx_flow;
 	struct mutex qrtr_tx_lock; /* for qrtr_tx_flow */
 
 	struct sk_buff_head rx_queue;
@@ -170,6 +170,7 @@ static void __qrtr_node_release(struct kref *kref)
 	struct qrtr_tx_flow *flow;
 	unsigned long flags;
 	void __rcu **slot;
+	unsigned long index;
 
 	spin_lock_irqsave(&qrtr_nodes_lock, flags);
 	/* If the node is a bridge for other nodes, there are possibly
@@ -187,11 +188,9 @@ static void __qrtr_node_release(struct kref *kref)
 	skb_queue_purge(&node->rx_queue);
 
 	/* Free tx flow counters */
-	radix_tree_for_each_slot(slot, &node->qrtr_tx_flow, &iter, 0) {
-		flow = *slot;
-		radix_tree_iter_delete(&node->qrtr_tx_flow, &iter, slot);
+	xa_for_each(&node->qrtr_tx_flow, index, flow)
 		kfree(flow);
-	}
+	xa_destroy(&node->qrtr_tx_flow);
 	kfree(node);
 }
 
@@ -226,9 +225,7 @@ static void qrtr_tx_resume(struct qrtr_node *node, struct sk_buff *skb)
 
 	key = remote_node << 32 | remote_port;
 
-	rcu_read_lock();
-	flow = radix_tree_lookup(&node->qrtr_tx_flow, key);
-	rcu_read_unlock();
+	flow = xa_load(&node->qrtr_tx_flow, key);
 	if (flow) {
 		spin_lock(&flow->resume_tx.lock);
 		flow->pending = 0;
@@ -267,12 +264,13 @@ static int qrtr_tx_wait(struct qrtr_node *node, int dest_node, int dest_port,
 		return 0;
 
 	mutex_lock(&node->qrtr_tx_lock);
-	flow = radix_tree_lookup(&node->qrtr_tx_flow, key);
+	flow = xa_load(&node->qrtr_tx_flow, key);
 	if (!flow) {
 		flow = kzalloc(sizeof(*flow), GFP_KERNEL);
 		if (flow) {
 			init_waitqueue_head(&flow->resume_tx);
-			if (radix_tree_insert(&node->qrtr_tx_flow, key, flow)) {
+			if (xa_err(xa_store(&node->qrtr_tx_flow, key, flow,
+					    GFP_KERNEL))) {
 				kfree(flow);
 				flow = NULL;
 			}
@@ -324,9 +322,7 @@ static void qrtr_tx_flow_failed(struct qrtr_node *node, int dest_node,
 	unsigned long key = (u64)dest_node << 32 | dest_port;
 	struct qrtr_tx_flow *flow;
 
-	rcu_read_lock();
-	flow = radix_tree_lookup(&node->qrtr_tx_flow, key);
-	rcu_read_unlock();
+	flow = xa_load(&node->qrtr_tx_flow, key);
 	if (flow) {
 		spin_lock_irq(&flow->resume_tx.lock);
 		flow->tx_failed = 1;
@@ -593,7 +589,7 @@ int qrtr_endpoint_register(struct qrtr_endpoint *ep, unsigned int nid)
 	node->nid = QRTR_EP_NID_AUTO;
 	node->ep = ep;
 
-	INIT_RADIX_TREE(&node->qrtr_tx_flow, GFP_KERNEL);
+	xa_init(&node->qrtr_tx_flow);
 	mutex_init(&node->qrtr_tx_lock);
 
 	qrtr_node_assign(node, nid);
@@ -621,6 +617,7 @@ void qrtr_endpoint_unregister(struct qrtr_endpoint *ep)
 	struct qrtr_tx_flow *flow;
 	struct sk_buff *skb;
 	unsigned long flags;
+	unsigned long index;
 	void __rcu **slot;
 
 	mutex_lock(&node->ep_lock);
@@ -643,10 +640,8 @@ void qrtr_endpoint_unregister(struct qrtr_endpoint *ep)
 
 	/* Wake up any transmitters waiting for resume-tx from the node */
 	mutex_lock(&node->qrtr_tx_lock);
-	radix_tree_for_each_slot(slot, &node->qrtr_tx_flow, &iter, 0) {
-		flow = *slot;
+	xa_for_each(&node->qrtr_tx_flow, index, flow)
 		wake_up_interruptible_all(&flow->resume_tx);
-	}
 	mutex_unlock(&node->qrtr_tx_lock);
 
 	qrtr_node_release(node);
Signature · 1 line
2.53.0
author maintainer
#349 lore raw

PATCH 325/491 net: ipv6: ndisc: fix ndisc_ra_useropt to initialize nduseropt_padX fields to zero to prevent an info-leak

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yochai Eisenrich <[email protected]>

[ Upstream commit ae05340ccaa9d347fe85415609e075545bec589f ]

When processing Router Advertisements with user options the kernel builds an RTM_NEWNDUSEROPT netlink message. The nduseroptmsg struct has three padding fields that are never zeroed and can leak kernel data

The fix is simple, just zeroes the padding fields.

Fixes 31910575a9de ("[IPv6]: Export userland ND options through netlink (RDNSS support)")
---
 net/ipv6/ndisc.c | 3 +++
 1 file changed, 3 insertions(+)
net/ipv6/ndisc.c +3 -0
index 43ad4e5db5941..de389b519700d 100644
@@ -1155,6 +1155,9 @@ static void ndisc_ra_useropt(struct sk_buff *ra, struct nd_opt_hdr *opt)
 	ndmsg->nduseropt_icmp_type = icmp6h->icmp6_type;
 	ndmsg->nduseropt_icmp_code = icmp6h->icmp6_code;
 	ndmsg->nduseropt_opts_len = opt->nd_opt_len << 3;
+	ndmsg->nduseropt_pad1 = 0;
+	ndmsg->nduseropt_pad2 = 0;
+	ndmsg->nduseropt_pad3 = 0;
 
 	memcpy(ndmsg + 1, opt, opt->nd_opt_len << 3);
 
Signature · 1 line
2.53.0
author maintainer
#350 lore raw

PATCH 326/491 tg3: Fix race for querying speed/duplex

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thomas Bogendoerfer <[email protected]>

[ Upstream commit bb417456c7814d1493d98b7dd9c040bf3ce3b4ed ]

When driver signals carrier up via netif_carrier_on() its internal link_up state isn't updated immediately. This leads to inconsistent speed/duplex in /proc/net/bonding/bondX where the speed and duplex is shown as unknown while ethtool shows correct values. Fix this by using netif_carrier_ok() for link checking in get_ksettings function.

Fixes 84421b99cedc ("tg3: Update link_up flag for phylib devices")
---
 drivers/net/ethernet/broadcom/tg3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/broadcom/tg3.c +1 -1
index e8be9e5a244fd..954a51fe0cd70 100644
@@ -12232,7 +12232,7 @@ static int tg3_get_link_ksettings(struct net_device *dev,
 	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising,
 						advertising);
 
-	if (netif_running(dev) && tp->link_up) {
+	if (netif_running(dev) && netif_carrier_ok(dev)) {
 		cmd->base.speed = tp->link_config.active_speed;
 		cmd->base.duplex = tp->link_config.active_duplex;
 		ethtool_convert_legacy_u32_to_link_mode(
Signature · 1 line
2.53.0
author maintainer
#351 lore raw

PATCH 327/491 ipv6: icmp: clear skb2->cb[] in ip6_err_gen_icmpv6_unreach()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 86ab3e55673a7a49a841838776f1ab18d23a67b5 ]

Sashiko AI-review observed:

In ip6_err_gen_icmpv6_unreach(), the skb is an outer IPv4 ICMP error packet where its cb contains an IPv4 inet_skb_parm. When skb is cloned into skb2 and passed to icmp6_send(), it uses IP6CB(skb2).

IP6CB interprets the IPv4 inet_skb_parm as an inet6_skb_parm. The cipso offset in inet_skb_parm.opt directly overlaps with dsthao in inet6_skb_parm at offset 18.

If an attacker sends a forged ICMPv4 error with a CIPSO IP option, dsthao would be a non-zero offset. Inside icmp6_send(), mip6_addr_swap() is called and uses ipv6_find_tlv(skb, opt->dsthao, IPV6_TLV_HAO).

This would scan the inner, attacker-controlled IPv6 packet starting at that offset, potentially returning a fake TLV without checking if the remaining packet length can hold the full 18-byte struct ipv6_destopt_hao.

Could mip6_addr_swap() then perform a 16-byte swap that extends past the end of the packet data into skb_shared_info?

Should the cb array also be cleared in ip6_err_gen_icmpv6_unreach() and ip6ip6_err() to prevent this?

This patch implements the first suggestion.

I am not sure if ip6ip6_err() needs to be changed. A separate patch would be better anyway.

---
 net/ipv6/icmp.c | 3 +++
 1 file changed, 3 insertions(+)
net/ipv6/icmp.c +3 -0
index d01165bb6a32b..65846f4451894 100644
@@ -673,6 +673,9 @@ int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type,
 	if (!skb2)
 		return 1;
 
+	/* Remove debris left by IPv4 stack. */
+	memset(IP6CB(skb2), 0, sizeof(*IP6CB(skb2)));
+
 	skb_dst_drop(skb2);
 	skb_pull(skb2, nhs);
 	skb_reset_network_header(skb2);
Signature · 1 line
2.53.0
author maintainer
#352 lore raw

PATCH 328/491 ip6_tunnel: clear skb2->cb[] in ip4ip6_err()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 2edfa31769a4add828a7e604b21cb82aaaa05925 ]

Oskar Kjos reported the following problem.

ip4ip6_err() calls icmp_send() on a cloned skb whose cb[] was written by the IPv6 receive path as struct inet6_skb_parm. icmp_send() passes IPCB(skb2) to __ip_options_echo(), which interprets that cb[] region as struct inet_skb_parm (IPv4). The layouts differ: inet6_skb_parm.nhoff at offset 14 overlaps inet_skb_parm.opt.rr, producing a non-zero rr value. __ip_options_echo() then reads optlen from attacker-controlled packet data at sptr[rr+1] and copies that many bytes into dopt->__data, a fixed 40-byte stack buffer (IP_OPTIONS_DATA_FIXED_SIZE).

To fix this we clear skb2->cb[], as suggested by Oskar Kjos.

Also add minimal IPv4 header validation (version == 4, ihl >= 5).

Fixes c4d3efafcc93 ("[IPV6] IP6TUNNEL: Add support to IPv4 over IPv6 tunnel.")
---
 net/ipv6/ip6_tunnel.c | 5 +++++
 1 file changed, 5 insertions(+)
net/ipv6/ip6_tunnel.c +5 -0
index e8c4e02e75d43..dda90c77f8984 100644
@@ -634,11 +634,16 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	if (!skb2)
 		return 0;
 
+	/* Remove debris left by IPv6 stack. */
+	memset(IPCB(skb2), 0, sizeof(*IPCB(skb2)));
+
 	skb_dst_drop(skb2);
 
 	skb_pull(skb2, offset);
 	skb_reset_network_header(skb2);
 	eiph = ip_hdr(skb2);
+	if (eiph->version != 4 || eiph->ihl < 5)
+		goto out;
 
 	/* Try to guess incoming interface */
 	rt = ip_route_output_ports(dev_net(skb->dev), &fl4, NULL, eiph->saddr,
Signature · 1 line
2.53.0
author maintainer
#353 lore raw

PATCH 329/491 bridge: br_nd_send: linearize skb before parsing ND options

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yang Yang <[email protected]>

[ Upstream commit a01aee7cafc575bb82f5529e8734e7052f9b16ea ]

br_nd_send() parses neighbour discovery options from ns->opt[] and assumes that these options are in the linear part of request.

Its callers only guarantee that the ICMPv6 header and target address are available, so the option area can still be non-linear. Parsing ns->opt[] in that case can access data past the linear buffer.

Linearize request before option parsing and derive ns from the linear network header.

---
 net/bridge/br_arp_nd_proxy.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
net/bridge/br_arp_nd_proxy.c +7 -5
index 3db1def4437b3..a44d14b94865e 100644
@@ -248,12 +248,12 @@ struct nd_msg *br_is_nd_neigh_msg(struct sk_buff *skb, struct nd_msg *msg)
 
 static void br_nd_send(struct net_bridge *br, struct net_bridge_port *p,
 		       struct sk_buff *request, struct neighbour *n,
-		       __be16 vlan_proto, u16 vlan_tci, struct nd_msg *ns)
+		       __be16 vlan_proto, u16 vlan_tci)
 {
 	struct net_device *dev = request->dev;
 	struct net_bridge_vlan_group *vg;
+	struct nd_msg *na, *ns;
 	struct sk_buff *reply;
-	struct nd_msg *na;
 	struct ipv6hdr *pip6;
 	int na_olen = 8; /* opt hdr + ETH_ALEN for target */
 	int ns_olen;
@@ -261,7 +261,7 @@ static void br_nd_send(struct net_bridge *br, struct net_bridge_port *p,
 	u8 *daddr;
 	u16 pvid;
 
-	if (!dev)
+	if (!dev || skb_linearize(request))
 		return;
 
 	len = LL_RESERVED_SPACE(dev) + sizeof(struct ipv6hdr) +
@@ -278,6 +278,8 @@ static void br_nd_send(struct net_bridge *br, struct net_bridge_port *p,
 	skb_set_mac_header(reply, 0);
 
 	daddr = eth_hdr(request)->h_source;
+	ns = (struct nd_msg *)(skb_network_header(request) +
+			       sizeof(struct ipv6hdr));
 
 	/* Do we need option processing ? */
 	ns_olen = request->len - (skb_network_offset(request) +
@@ -465,9 +467,9 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br,
 				if (vid != 0)
 					br_nd_send(br, p, skb, n,
 						   skb->vlan_proto,
-						   skb_vlan_tag_get(skb), msg);
+						   skb_vlan_tag_get(skb));
 				else
-					br_nd_send(br, p, skb, n, 0, 0, msg);
+					br_nd_send(br, p, skb, n, 0, 0);
 				replied = true;
 			}
 
Signature · 1 line
2.53.0
author maintainer
#354 lore raw

PATCH 330/491 net/sched: sch_hfsc: fix divide-by-zero in rtsc_min()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Xiang Mei <[email protected]>

[ Upstream commit 4576100b8cd03118267513cafacde164b498b322 ]

m2sm() converts a u32 slope to a u64 scaled value. For large inputs (e.g. m1=4000000000), the result can reach 2^32. rtsc_min() stores the difference of two such u64 values in a u32 variable dsm and uses it as a divisor. When the difference is exactly 2^32 the truncation yields zero, causing a divide-by-zero oops in the concave-curve intersection path:

  Oops: divide error: 0000
  RIP: 0010:rtsc_min (net/sched/sch_hfsc.c:601)
  Call Trace:
   init_ed (net/sched/sch_hfsc.c:629)
   hfsc_enqueue (net/sched/sch_hfsc.c:1569)
   [...]

Widen dsm to u64 and replace do_div() with div64_u64() so the full difference is preserved.

---
 net/sched/sch_hfsc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/sched/sch_hfsc.c +2 -2
index 25360061ad288..3a271ad16443b 100644
@@ -556,7 +556,7 @@ static void
 rtsc_min(struct runtime_sc *rtsc, struct internal_sc *isc, u64 x, u64 y)
 {
 	u64 y1, y2, dx, dy;
-	u32 dsm;
+	u64 dsm;
 
 	if (isc->sm1 <= isc->sm2) {
 		/* service curve is convex */
@@ -599,7 +599,7 @@ rtsc_min(struct runtime_sc *rtsc, struct internal_sc *isc, u64 x, u64 y)
 	 */
 	dx = (y1 - y) << SM_SHIFT;
 	dsm = isc->sm1 - isc->sm2;
-	do_div(dx, dsm);
+	dx = div64_u64(dx, dsm);
 	/*
 	 * check if (x, y1) belongs to the 1st segment of rtsc.
 	 * if so, add the offset.
Signature · 1 line
2.53.0
author maintainer
#355 lore raw

PATCH 331/491 ipv6: prevent possible UaF in addrconf_permanent_addr()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paolo Abeni <[email protected]>

[ Upstream commit fd63f185979b047fb22a0dfc6bd94d0cab6a6a70 ]

The mentioned helper try to warn the user about an exceptional condition, but the message is delivered too late, accessing the ipv6 after its possible deletion.

Reorder the statement to avoid the possible UaF; while at it, place the warning outside the idev->lock as it needs no protection.

---
 net/ipv6/addrconf.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
net/ipv6/addrconf.c +3 -3
index ced20abf4ef8e..758cea239e618 100644
@@ -3509,12 +3509,12 @@ static void addrconf_permanent_addr(struct net *net, struct net_device *dev)
 		if ((ifp->flags & IFA_F_PERMANENT) &&
 		    fixup_permanent_addr(net, idev, ifp) < 0) {
 			write_unlock_bh(&idev->lock);
-			in6_ifa_hold(ifp);
-			ipv6_del_addr(ifp);
-			write_lock_bh(&idev->lock);
 
 			net_info_ratelimited("%s: Failed to add prefix route for address %pI6c; dropping\n",
 					     idev->dev->name, &ifp->addr);
+			in6_ifa_hold(ifp);
+			ipv6_del_addr(ifp);
+			write_lock_bh(&idev->lock);
 		}
 	}
 
Signature · 1 line
2.53.0
author maintainer
#356 lore raw

PATCH 332/491 net: sched: cls_api: fix tc_chain_fill_node to initialize tcm_info to zero to prevent an info-leak

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yochai Eisenrich <[email protected]>

[ Upstream commit e6e3eb5ee89ac4c163d46429391c889a1bb5e404 ]

When building netlink messages, tc_chain_fill_node() never initializes the tcm_info field of struct tcmsg. Since the allocation is not zeroed, kernel heap memory is leaked to userspace through this 4-byte field.

The fix simply zeroes tcm_info alongside the other fields that are already initialized.

---
 net/sched/cls_api.c | 1 +
 1 file changed, 1 insertion(+)
net/sched/cls_api.c +1 -0
index beedd0d2b5097..27847fae053d3 100644
@@ -2669,6 +2669,7 @@ static int tc_chain_fill_node(const struct tcf_proto_ops *tmplt_ops,
 	tcm->tcm__pad1 = 0;
 	tcm->tcm__pad2 = 0;
 	tcm->tcm_handle = 0;
+	tcm->tcm_info = 0;
 	if (block->q) {
 		tcm->tcm_ifindex = qdisc_dev(block->q)->ifindex;
 		tcm->tcm_parent = block->q->handle;
Signature · 1 line
2.53.0
author maintainer
#357 lore raw

PATCH 333/491 NFC: pn533: bound the UART receive buffer

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pengpeng Hou <[email protected]>

[ Upstream commit 30fe3f5f6494f827d812ff179f295a8e532709d6 ]

pn532_receive_buf() appends every incoming byte to dev->recv_skb and only resets the buffer after pn532_uart_rx_is_frame() recognizes a complete frame. A continuous stream of bytes without a valid PN532 frame header therefore keeps growing the skb until skb_put_u8() hits the tail limit.

Drop the accumulated partial frame once the fixed receive buffer is full so malformed UART traffic cannot grow the skb past PN532_UART_SKB_BUFF_LEN.

Fixes c656aa4c27b1 ("nfc: pn533: add UART phy driver")
---
 drivers/nfc/pn533/uart.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/nfc/pn533/uart.c +3 -0
index e92535ebb5287..7465b4b48c829 100644
@@ -211,6 +211,9 @@ static int pn532_receive_buf(struct serdev_device *serdev,
 
 	del_timer(&dev->cmd_timeout);
 	for (i = 0; i < count; i++) {
+		if (unlikely(!skb_tailroom(dev->recv_skb)))
+			skb_trim(dev->recv_skb, 0);
+
 		skb_put_u8(dev->recv_skb, *data++);
 		if (!pn532_uart_rx_is_frame(dev->recv_skb))
 			continue;
Signature · 1 line
2.53.0
author maintainer
#358 lore raw

PATCH 334/491 net: xilinx: axienet: Correct BD length masks to match AXIDMA IP spec

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Suraj Gupta <[email protected]>

[ Upstream commit 393e0b4f178ec7fce1141dacc3304e3607a92ee9 ]

The XAXIDMA_BD_CTRL_LENGTH_MASK and XAXIDMA_BD_STS_ACTUAL_LEN_MASK macros were defined as 0x007FFFFF (23 bits), but the AXI DMA IP product guide (PG021) specifies the buffer length field as bits 25:0 (26 bits). Update both masks to match the IP documentation.

In practice this had no functional impact, since Ethernet frames are far smaller than 2^23 bytes and the extra bits were always zero, but the masks should still reflect the hardware specification.

Fixes 8a3b7a252dca ("drivers/net/ethernet/xilinx: added Xilinx AXI Ethernet driver")
---
 drivers/net/ethernet/xilinx/xilinx_axienet.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/xilinx/xilinx_axienet.h +2 -2
index 071822028ea5e..516f40e7560ac 100644
@@ -103,7 +103,7 @@
 #define XAXIDMA_BD_HAS_DRE_MASK		0xF00 /* Whether has DRE mask */
 #define XAXIDMA_BD_WORDLEN_MASK		0xFF /* Whether has DRE mask */
 
-#define XAXIDMA_BD_CTRL_LENGTH_MASK	0x007FFFFF /* Requested len */
+#define XAXIDMA_BD_CTRL_LENGTH_MASK	GENMASK(25, 0) /* Requested len */
 #define XAXIDMA_BD_CTRL_TXSOF_MASK	0x08000000 /* First tx packet */
 #define XAXIDMA_BD_CTRL_TXEOF_MASK	0x04000000 /* Last tx packet */
 #define XAXIDMA_BD_CTRL_ALL_MASK	0x0C000000 /* All control bits */
@@ -129,7 +129,7 @@
 #define XAXIDMA_BD_CTRL_TXEOF_MASK	0x04000000 /* Last tx packet */
 #define XAXIDMA_BD_CTRL_ALL_MASK	0x0C000000 /* All control bits */
 
-#define XAXIDMA_BD_STS_ACTUAL_LEN_MASK	0x007FFFFF /* Actual len */
+#define XAXIDMA_BD_STS_ACTUAL_LEN_MASK	GENMASK(25, 0) /* Actual len */
 #define XAXIDMA_BD_STS_COMPLETE_MASK	0x80000000 /* Completed */
 #define XAXIDMA_BD_STS_DEC_ERR_MASK	0x40000000 /* Decode error */
 #define XAXIDMA_BD_STS_SLV_ERR_MASK	0x20000000 /* Slave error */
Signature · 1 line
2.53.0
author maintainer
#359 lore raw

PATCH 335/491 bpf: Fix regsafe() for pointers to packet

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alexei Starovoitov <[email protected]>

[ Upstream commit a8502a79e832b861e99218cbd2d8f4312d62e225 ]

In case rold->reg->range == BEYOND_PKT_END && rcur->reg->range == N regsafe() may return true which may lead to current state with valid packet range not being explored. Fix the bug.

---
 kernel/bpf/verifier.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
kernel/bpf/verifier.c +6 -1
index a2ccbd97e8dd2..e6cf29b117e1a 100644
@@ -9459,8 +9459,13 @@ static bool regsafe(struct bpf_verifier_env *env, struct bpf_reg_state *rold,
 		 * since someone could have accessed through (ptr - k), or
 		 * even done ptr -= k in a register, to get a safe access.
 		 */
-		if (rold->range > rcur->range)
+		if (rold->range < 0 || rcur->range < 0) {
+			/* special case for [BEYOND|AT]_PKT_END */
+			if (rold->range != rcur->range)
+				return false;
+		} else if (rold->range > rcur->range) {
 			return false;
+		}
 		/* If the offsets don't match, we can't trust our alignment;
 		 * nor can we be sure that we won't fall out of range.
 		 */
Signature · 1 line
2.53.0
author maintainer
#360 lore raw

PATCH 336/491 net: ipv6: flowlabel: defer exclusive option free until RCU teardown

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhengchuan Liang <[email protected]>

[ Upstream commit 9ca562bb8e66978b53028fa32b1a190708e6a091 ]

ip6fl_seq_show() walks the global flowlabel hash under the seq-file RCU read-side lock and prints fl->opt->opt_nflen when an option block is present.

Exclusive flowlabels currently free fl->opt as soon as fl->users drops to zero in fl_release(). However, the surrounding struct ip6_flowlabel remains visible in the global hash table until later garbage collection removes it and fl_free_rcu() finally tears it down.

A concurrent /proc/net/ip6_flowlabel reader can therefore race that early kfree() and dereference freed option state, triggering a crash in ip6fl_seq_show().

Fix this by keeping fl->opt alive until fl_free_rcu(). That matches the lifetime already required for the enclosing flowlabel while readers can still reach it under RCU.

---
 net/ipv6/ip6_flowlabel.c | 5 -----
 1 file changed, 5 deletions(-)
net/ipv6/ip6_flowlabel.c +0 -5
index ceb85c67ce395..bb528d0ddb73a 100644
@@ -133,11 +133,6 @@ static void fl_release(struct ip6_flowlabel *fl)
 		if (time_after(ttd, fl->expires))
 			fl->expires = ttd;
 		ttd = fl->expires;
-		if (fl->opt && fl->share == IPV6_FL_S_EXCL) {
-			struct ipv6_txoptions *opt = fl->opt;
-			fl->opt = NULL;
-			kfree(opt);
-		}
 		if (!timer_pending(&ip6_fl_gc_timer) ||
 		    time_after(ip6_fl_gc_timer.expires, ttd))
 			mod_timer(&ip6_fl_gc_timer, ttd);
Signature · 1 line
2.53.0
author maintainer
#361 lore raw

PATCH 337/491 netfilter: nfnetlink_log: account for netlink header size

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

[ Upstream commit 6d52a4a0520a6696bdde51caa11f2d6821cd0c01 ]

This is a followup to an old bug fix: NLMSG_DONE needs to account for the netlink header size, not just the attribute size.

This can result in a WARN splat + drop of the netlink message, but other than this there are no ill effects.

Fixes 9dfa1dfe4d5e ("netfilter: nf_log: account for size of NLMSG_DONE attribute")
---
 net/netfilter/nfnetlink_log.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/netfilter/nfnetlink_log.c +1 -1
index 3a71ba47224c4..d41560d4812d0 100644
@@ -711,7 +711,7 @@ nfulnl_log_packet(struct net *net,
 		+ nla_total_size(plen)			/* prefix */
 		+ nla_total_size(sizeof(struct nfulnl_msg_packet_hw))
 		+ nla_total_size(sizeof(struct nfulnl_msg_packet_timestamp))
-		+ nla_total_size(sizeof(struct nfgenmsg));	/* NLMSG_DONE */
+		+ nlmsg_total_size(sizeof(struct nfgenmsg));	/* NLMSG_DONE */
 
 	if (in && skb_mac_header_was_set(skb)) {
 		size += nla_total_size(skb->dev->hard_header_len)
Signature · 1 line
2.53.0
author maintainer
#362 lore raw

PATCH 338/491 netfilter: x_tables: ensure names are nul-terminated

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

[ Upstream commit a958a4f90ddd7de0800b33ca9d7b886b7d40f74e ]

Reject names that lack a \0 character before feeding them to functions that expect c-strings.

Fixes tag is the most recent commit that needs this change.

Fixes c38c4597e4bf ("netfilter: implement xt_cgroup cgroup2 path match")
---
 net/netfilter/xt_cgroup.c  | 6 ++++++
 net/netfilter/xt_rateest.c | 5 +++++
 2 files changed, 11 insertions(+)
net/netfilter/xt_cgroup.c +6 -0
index c0f5e9a4f3c65..bfc98719684e2 100644
@@ -53,6 +53,9 @@ static int cgroup_mt_check_v1(const struct xt_mtchk_param *par)
 
 	info->priv = NULL;
 	if (info->has_path) {
+		if (strnlen(info->path, sizeof(info->path)) >= sizeof(info->path))
+			return -ENAMETOOLONG;
+
 		cgrp = cgroup_get_from_path(info->path);
 		if (IS_ERR(cgrp)) {
 			pr_info_ratelimited("invalid path, errno=%ld\n",
@@ -85,6 +88,9 @@ static int cgroup_mt_check_v2(const struct xt_mtchk_param *par)
 
 	info->priv = NULL;
 	if (info->has_path) {
+		if (strnlen(info->path, sizeof(info->path)) >= sizeof(info->path))
+			return -ENAMETOOLONG;
+
 		cgrp = cgroup_get_from_path(info->path);
 		if (IS_ERR(cgrp)) {
 			pr_info_ratelimited("invalid path, errno=%ld\n",
net/netfilter/xt_rateest.c +5 -0
index 72324bd976af8..b1d736c15fcbe 100644
@@ -91,6 +91,11 @@ static int xt_rateest_mt_checkentry(const struct xt_mtchk_param *par)
 		goto err1;
 	}
 
+	if (strnlen(info->name1, sizeof(info->name1)) >= sizeof(info->name1))
+		return -ENAMETOOLONG;
+	if (strnlen(info->name2, sizeof(info->name2)) >= sizeof(info->name2))
+		return -ENAMETOOLONG;
+
 	ret  = -ENOENT;
 	est1 = xt_rateest_lookup(par->net, info->name1);
 	if (!est1)
Signature · 1 line
2.53.0
author maintainer
#363 lore raw

PATCH 339/491 netfilter: ipset: use nla_strcmp for IPSET_ATTR_NAME attr

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

[ Upstream commit b7e8590987aa94c9dc51518fad0e58cb887b1db5 ]

IPSET_ATTR_NAME and IPSET_ATTR_NAMEREF are of NLA_STRING type, they cannot be treated like a c-string.

They either have to be switched to NLA_NUL_STRING, or the compare operations need to use the nla functions.

Fixes f830837f0eed ("netfilter: ipset: list:set set type support")
---
 include/linux/netfilter/ipset/ip_set.h | 2 +-
 net/netfilter/ipset/ip_set_core.c      | 4 ++--
 net/netfilter/ipset/ip_set_list_set.c  | 4 ++--
 3 files changed, 5 insertions(+), 5 deletions(-)
include/linux/netfilter/ipset/ip_set.h +1 -1
index f27894e50ef19..4269b64252787 100644
@@ -306,7 +306,7 @@ enum {
 
 /* register and unregister set references */
 extern ip_set_id_t ip_set_get_byname(struct net *net,
-				     const char *name, struct ip_set **set);
+				     const struct nlattr *name, struct ip_set **set);
 extern void ip_set_put_byindex(struct net *net, ip_set_id_t index);
 extern void ip_set_name_byindex(struct net *net, ip_set_id_t index, char *name);
 extern ip_set_id_t ip_set_nfnl_get_byindex(struct net *net, ip_set_id_t index);
net/netfilter/ipset/ip_set_core.c +2 -2
index a265efd31ba96..cf827d72581e1 100644
@@ -823,7 +823,7 @@ EXPORT_SYMBOL_GPL(ip_set_del);
  *
  */
 ip_set_id_t
-ip_set_get_byname(struct net *net, const char *name, struct ip_set **set)
+ip_set_get_byname(struct net *net, const struct nlattr *name, struct ip_set **set)
 {
 	ip_set_id_t i, index = IPSET_INVALID_ID;
 	struct ip_set *s;
@@ -832,7 +832,7 @@ ip_set_get_byname(struct net *net, const char *name, struct ip_set **set)
 	rcu_read_lock();
 	for (i = 0; i < inst->ip_set_max; i++) {
 		s = rcu_dereference(inst->ip_set_list)[i];
-		if (s && STRNCMP(s->name, name)) {
+		if (s && nla_strcmp(name, s->name) == 0) {
 			__ip_set_get(s);
 			index = i;
 			*set = s;
net/netfilter/ipset/ip_set_list_set.c +2 -2
index 5cc35b553a048..7d1ba6ad514f5 100644
@@ -367,7 +367,7 @@ list_set_uadt(struct ip_set *set, struct nlattr *tb[],
 	ret = ip_set_get_extensions(set, tb, &ext);
 	if (ret)
 		return ret;
-	e.id = ip_set_get_byname(map->net, nla_data(tb[IPSET_ATTR_NAME]), &s);
+	e.id = ip_set_get_byname(map->net, tb[IPSET_ATTR_NAME], &s);
 	if (e.id == IPSET_INVALID_ID)
 		return -IPSET_ERR_NAME;
 	/* "Loop detection" */
@@ -389,7 +389,7 @@ list_set_uadt(struct ip_set *set, struct nlattr *tb[],
 
 	if (tb[IPSET_ATTR_NAMEREF]) {
 		e.refid = ip_set_get_byname(map->net,
-					    nla_data(tb[IPSET_ATTR_NAMEREF]),
+					    tb[IPSET_ATTR_NAMEREF],
 					    &s);
 		if (e.refid == IPSET_INVALID_ID) {
 			ret = -IPSET_ERR_NAMEREF;
Signature · 1 line
2.53.0
author maintainer
#364 lore raw

PATCH 340/491 netfilter: nf_conntrack_helper: pass helper to expect cleanup

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Qi Tang <[email protected]>

[ Upstream commit a242a9ae58aa46ff7dae51ce64150a93957abe65 ]

nf_conntrack_helper_unregister() calls nf_ct_expect_iterate_destroy() to remove expectations belonging to the helper being unregistered. However, it passes NULL instead of the helper pointer as the data argument, so expect_iter_me() never matches any expectation and all of them survive the cleanup.

After unregister returns, nfnl_cthelper_del() frees the helper object immediately. Subsequent expectation dumps or packet-driven init_conntrack() calls then dereference the freed exp->helper, causing a use-after-free.

Pass the actual helper pointer so expectations referencing it are properly destroyed before the helper object is freed.

  BUG: KASAN: slab-use-after-free in string+0x38f/0x430
  Read of size 1 at addr ffff888003b14d20 by task poc/103
  Call Trace:
   string+0x38f/0x430
   vsnprintf+0x3cc/0x1170
   seq_printf+0x17a/0x240
   exp_seq_show+0x2e5/0x560
   seq_read_iter+0x419/0x1280
   proc_reg_read+0x1ac/0x270
   vfs_read+0x179/0x930
   ksys_read+0xef/0x1c0
  Freed by task 103:
  The buggy address is located 32 bytes inside of
   freed 192-byte region [ffff888003b14d00, ffff888003b14dc0)
Fixes ac7b84839003 ("netfilter: expect: add and use nf_ct_expect_iterate helpers")
---
 net/netfilter/nf_conntrack_helper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/netfilter/nf_conntrack_helper.c +1 -1
index 89174c91053ed..24f3f8d5e6998 100644
@@ -468,7 +468,7 @@ void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me)
 	 */
 	synchronize_rcu();
 
-	nf_ct_expect_iterate_destroy(expect_iter_me, NULL);
+	nf_ct_expect_iterate_destroy(expect_iter_me, me);
 	nf_ct_iterate_destroy(unhelp, me);
 
 	/* Maybe someone has gotten the helper already when unhelp above.
Signature · 1 line
2.53.0
author maintainer
#365 lore raw

PATCH 341/491 netfilter: ctnetlink: zero expect NAT fields when CTA_EXPECT_NAT absent

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Qi Tang <[email protected]>

[ Upstream commit 35177c6877134a21315f37d57a5577846225623e ]

ctnetlink_alloc_expect() allocates expectations from a non-zeroing slab cache via nf_ct_expect_alloc(). When CTA_EXPECT_NAT is not present in the netlink message, saved_addr and saved_proto are never initialized. Stale data from a previous slab occupant can then be dumped to userspace by ctnetlink_exp_dump_expect(), which checks these fields to decide whether to emit CTA_EXPECT_NAT.

The safe sibling nf_ct_expect_init(), used by the packet path, explicitly zeroes these fields.

Zero saved_addr, saved_proto and dir in the else branch, guarded by IS_ENABLED(CONFIG_NF_NAT) since these fields only exist when NAT is enabled.

Confirmed by priming the expect slab with NAT-bearing expectations, freeing them, creating a new expectation without CTA_EXPECT_NAT, and observing that the ctnetlink dump emits a spurious CTA_EXPECT_NAT containing stale data from the prior allocation.

Fixes 076a0ca02644 ("netfilter: ctnetlink: add NAT support for expectations")
---
 net/netfilter/nf_conntrack_netlink.c | 6 ++++++
 1 file changed, 6 insertions(+)
net/netfilter/nf_conntrack_netlink.c +6 -0
index 8d35e17a0f7e7..befc9d2bc0b52 100644
@@ -3568,6 +3568,12 @@ ctnetlink_alloc_expect(const struct nlattr * const cda[], struct nf_conn *ct,
 						 exp, nf_ct_l3num(ct));
 		if (err < 0)
 			goto err_out;
+#if IS_ENABLED(CONFIG_NF_NAT)
+	} else {
+		memset(&exp->saved_addr, 0, sizeof(exp->saved_addr));
+		memset(&exp->saved_proto, 0, sizeof(exp->saved_proto));
+		exp->dir = 0;
+#endif
 	}
 	return exp;
 err_out:
Signature · 1 line
2.53.0
author maintainer
#366 lore raw

PATCH 342/491 netfilter: x_tables: restrict xt_check_match/xt_check_target extensions for NFPROTO_ARP

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit 3d5d488f11776738deab9da336038add95d342d1 ]

Weiming Shi says:

xt_match and xt_target structs registered with NFPROTO_UNSPEC can be loaded by any protocol family through nft_compat. When such a match/target sets .hooks to restrict which hooks it may run on, the bitmask uses NF_INET_* constants. This is only correct for families whose hook layout matches NF_INET_*: IPv4, IPv6, INET, and bridge all share the same five hooks (PRE_ROUTING ... POST_ROUTING).

ARP only has three hooks (IN=0, OUT=1, FORWARD=2) with different semantics. Because NF_ARP_OUT == 1 == NF_INET_LOCAL_IN, the .hooks validation silently passes for the wrong reasons, allowing matches to run on ARP chains where the hook assumptions (e.g. state->in being set on input hooks) do not hold. This leads to NULL pointer dereferences; xt_devgroup is one concrete example:

 Oops: general protection fault, probably for non-canonical address 0xdffffc0000000044: 0000 [#1] SMP KASAN NOPTI
 KASAN: null-ptr-deref in range [0x0000000000000220-0x0000000000000227]
 RIP: 0010:devgroup_mt+0xff/0x350
 Call Trace:
  <TASK>
  nft_match_eval (net/netfilter/nft_compat.c:407)
  nft_do_chain (net/netfilter/nf_tables_core.c:285)
  nft_do_chain_arp (net/netfilter/nft_chain_filter.c:61)
  nf_hook_slow (net/netfilter/core.c:623)
  arp_xmit (net/ipv4/arp.c:666)
  </TASK>
 Kernel panic - not syncing: Fatal exception in interrupt

Fix it by restricting arptables to NFPROTO_ARP extensions only. Note that arptables-legacy only supports:

- arpt_CLASSIFY
- arpt_mangle
- arpt_MARK

that provide explicit NFPROTO_ARP match/target declarations.

Fixes 9291747f118d ("netfilter: xtables: add device group match")
---
 net/netfilter/x_tables.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)
net/netfilter/x_tables.c +23 -0
index 92e9d4ebc5e8d..94778fae2d91d 100644
@@ -481,6 +481,17 @@ int xt_check_match(struct xt_mtchk_param *par,
 				    par->match->table, par->table);
 		return -EINVAL;
 	}
+
+	/* NFPROTO_UNSPEC implies NF_INET_* hooks which do not overlap with
+	 * NF_ARP_IN,OUT,FORWARD, allow explicit extensions with NFPROTO_ARP
+	 * support.
+	 */
+	if (par->family == NFPROTO_ARP &&
+	    par->match->family != NFPROTO_ARP) {
+		pr_info_ratelimited("%s_tables: %s match: not valid for this family\n",
+				    xt_prefix[par->family], par->match->name);
+		return -EINVAL;
+	}
 	if (par->match->hooks && (par->hook_mask & ~par->match->hooks) != 0) {
 		char used[64], allow[64];
 
@@ -996,6 +1007,18 @@ int xt_check_target(struct xt_tgchk_param *par,
 				    par->target->table, par->table);
 		return -EINVAL;
 	}
+
+	/* NFPROTO_UNSPEC implies NF_INET_* hooks which do not overlap with
+	 * NF_ARP_IN,OUT,FORWARD, allow explicit extensions with NFPROTO_ARP
+	 * support.
+	 */
+	if (par->family == NFPROTO_ARP &&
+	    par->target->family != NFPROTO_ARP) {
+		pr_info_ratelimited("%s_tables: %s target: not valid for this family\n",
+				    xt_prefix[par->family], par->target->name);
+		return -EINVAL;
+	}
+
 	if (par->target->hooks && (par->hook_mask & ~par->target->hooks) != 0) {
 		char used[64], allow[64];
 
Signature · 1 line
2.53.0
author maintainer
#367 lore raw

PATCH 343/491 netfilter: nf_tables: reject immediate NF_QUEUE verdict

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit da107398cbd4bbdb6bffecb2ce86d5c9384f4cec ]

nft_queue is always used from userspace nftables to deliver the NF_QUEUE verdict. Immediately emitting an NF_QUEUE verdict is never used by the userspace nft tools, so reject immediate NF_QUEUE verdicts.

The arp family does not provide queue support, but such an immediate verdict is still reachable. Globally reject NF_QUEUE immediate verdicts to address this issue.

Fixes f342de4e2f33 ("netfilter: nf_tables: reject QUEUE/DROP verdict parameters")
---
 net/netfilter/nf_tables_api.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
net/netfilter/nf_tables_api.c +5 -2
index 189997bcaeef5..15486d3051f6b 100644
@@ -9362,8 +9362,6 @@ static int nft_verdict_init(const struct nft_ctx *ctx, struct nft_data *data,
 	switch (data->verdict.code) {
 	case NF_ACCEPT:
 	case NF_DROP:
-	case NF_QUEUE:
-		break;
 	case NFT_CONTINUE:
 	case NFT_BREAK:
 	case NFT_RETURN:
@@ -9398,6 +9396,11 @@ static int nft_verdict_init(const struct nft_ctx *ctx, struct nft_data *data,
 
 		data->verdict.chain = chain;
 		break;
+	case NF_QUEUE:
+		/* The nft_queue expression is used for this purpose, an
+		 * immediate NF_QUEUE verdict should not ever be seen here.
+		 */
+		fallthrough;
 	default:
 		return -EINVAL;
 	}
Signature · 1 line
2.53.0
author maintainer
#368 lore raw

PATCH 344/491 Bluetooth: MGMT: validate LTK enc_size on load

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Keenan Dong <[email protected]>

[ Upstream commit b8dbe9648d69059cfe3a28917bfbf7e61efd7f15 ]

Load Long Term Keys stores the user-provided enc_size and later uses it to size fixed-size stack operations when replying to LE LTK requests. An enc_size larger than the 16-byte key buffer can therefore overflow the reply stack buffer.

Reject oversized enc_size values while validating the management LTK record so invalid keys never reach the stored key state.

Fixes 346af67b8d11 ("Bluetooth: Add MGMT handlers for dealing with SMP LTK's")
---
 net/bluetooth/mgmt.c | 3 +++
 1 file changed, 3 insertions(+)
net/bluetooth/mgmt.c +3 -0
index fda0bd990dcb7..b768abbf2b121 100644
@@ -5948,6 +5948,9 @@ static bool ltk_is_valid(struct mgmt_ltk_info *key)
 	if (key->initiator != 0x00 && key->initiator != 0x01)
 		return false;
 
+	if (key->enc_size > sizeof(key->val))
+		return false;
+
 	switch (key->addr.type) {
 	case BDADDR_LE_PUBLIC:
 		return true;
Signature · 1 line
2.53.0
author maintainer
#369 lore raw

PATCH 345/491 rds: ib: reject FRMR registration before IB connection is established

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Weiming Shi <[email protected]>

[ Upstream commit a54ecccfae62c5c85259ae5ea5d9c20009519049 ]

rds_ib_get_mr() extracts the rds_ib_connection from conn->c_transport_data and passes it to rds_ib_reg_frmr() for FRWR memory registration. On a fresh outgoing connection, ic is allocated in rds_ib_conn_alloc() with i_cm_id = NULL because the connection worker has not yet called rds_ib_conn_path_connect() to create the rdma_cm_id. When sendmsg() with RDS_CMSG_RDMA_MAP is called on such a connection, the sendmsg path parses the control message before any connection establishment, allowing rds_ib_post_reg_frmr() to dereference ic->i_cm_id->qp and crash the kernel.

The existing guard in rds_ib_reg_frmr() only checks for !ic (added in commit 9e630bcb7701), which does not catch this case since ic is allocated early and is always non-NULL once the connection object exists.

 KASAN: null-ptr-deref in range [0x0000000000000010-0x0000000000000017]
 RIP: 0010:rds_ib_post_reg_frmr+0x50e/0x920
 Call Trace:
  rds_ib_post_reg_frmr (net/rds/ib_frmr.c:167)
  rds_ib_map_frmr (net/rds/ib_frmr.c:252)
  rds_ib_reg_frmr (net/rds/ib_frmr.c:430)
  rds_ib_get_mr (net/rds/ib_rdma.c:615)
  __rds_rdma_map (net/rds/rdma.c:295)
  rds_cmsg_rdma_map (net/rds/rdma.c:860)
  rds_sendmsg (net/rds/send.c:1363)
  ____sys_sendmsg
  do_syscall_64

Add a check in rds_ib_get_mr() that verifies ic, i_cm_id, and qp are all non-NULL before proceeding with FRMR registration, mirroring the guard already present in rds_ib_post_inv(). Return -ENODEV when the connection is not ready, which the existing error handling in rds_cmsg_send() converts to -EAGAIN for userspace retry and triggers rds_conn_connect_if_down() to start the connection worker.

Fixes 1659185fb4d0 ("RDS: IB: Support Fastreg MR (FRMR) memory registration mode")
---
 net/rds/ib_rdma.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
net/rds/ib_rdma.c +6 -1
index 8f070ee7e7426..30fca2169aa7a 100644
@@ -608,8 +608,13 @@ void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
 		return ibmr;
 	}
 
-	if (conn)
+	if (conn) {
 		ic = conn->c_transport_data;
+		if (!ic || !ic->i_cm_id || !ic->i_cm_id->qp) {
+			ret = -ENODEV;
+			goto out;
+		}
+	}
 
 	if (!rds_ibdev->mr_8k_pool || !rds_ibdev->mr_1m_pool) {
 		ret = -ENODEV;
Signature · 1 line
2.53.0
author maintainer
#370 lore raw

PATCH 346/491 net: macb: fix clk handling on PCI glue driver removal

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Fedor Pchelkin <[email protected]>

[ Upstream commit ce8fe5287b87e24e225c342f3b0ec04f0b3680fe ]

platform_device_unregister() may still want to use the registered clks during runtime resume callback.

Note that there is a commit d82d5303c4c5 ("net: macb: fix use after free on rmmod") that addressed the similar problem of clk vs platform device unregistration but just moved the bug to another place.

Save the pointers to clks into local variables for reuse after platform device is unregistered.

BUG: KASAN: use-after-free in clk_prepare+0x5a/0x60
Read of size 8 at addr ffff888104f85e00 by task modprobe/597

CPU: 2 PID: 597 Comm: modprobe Not tainted 6.1.164+ #114
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.1-0-g3208b098f51a-prebuilt.qemu.org 04/01/2014
Call Trace:
 <TASK>
 dump_stack_lvl+0x8d/0xba
 print_report+0x17f/0x496
 kasan_report+0xd9/0x180
 clk_prepare+0x5a/0x60
 macb_runtime_resume+0x13d/0x410 [macb]
 pm_generic_runtime_resume+0x97/0xd0
 __rpm_callback+0xc8/0x4d0
 rpm_callback+0xf6/0x230
 rpm_resume+0xeeb/0x1a70
 __pm_runtime_resume+0xb4/0x170
 bus_remove_device+0x2e3/0x4b0
 device_del+0x5b3/0xdc0
 platform_device_del+0x4e/0x280
 platform_device_unregister+0x11/0x50
 pci_device_remove+0xae/0x210
 device_remove+0xcb/0x180
 device_release_driver_internal+0x529/0x770
 driver_detach+0xd4/0x1a0
 bus_remove_driver+0x135/0x260
 driver_unregister+0x72/0xb0
 pci_unregister_driver+0x26/0x220
 __do_sys_delete_module+0x32e/0x550
 do_syscall_64+0x35/0x80
 entry_SYSCALL_64_after_hwframe+0x6e/0xd8
 </TASK>

Allocated by task 519:
 kasan_save_stack+0x2c/0x50
 kasan_set_track+0x21/0x30
 __kasan_kmalloc+0x8e/0x90
 __clk_register+0x458/0x2890
 clk_hw_register+0x1a/0x60
 __clk_hw_register_fixed_rate+0x255/0x410
 clk_register_fixed_rate+0x3c/0xa0
 macb_probe+0x1d8/0x42e [macb_pci]
 local_pci_probe+0xd7/0x190
 pci_device_probe+0x252/0x600
 really_probe+0x255/0x7f0
 __driver_probe_device+0x1ee/0x330
 driver_probe_device+0x4c/0x1f0
 __driver_attach+0x1df/0x4e0
 bus_for_each_dev+0x15d/0x1f0
 bus_add_driver+0x486/0x5e0
 driver_register+0x23a/0x3d0
 do_one_initcall+0xfd/0x4d0
 do_init_module+0x18b/0x5a0
 load_module+0x5663/0x7950
 __do_sys_finit_module+0x101/0x180
 do_syscall_64+0x35/0x80
 entry_SYSCALL_64_after_hwframe+0x6e/0xd8

Freed by task 597:
 kasan_save_stack+0x2c/0x50
 kasan_set_track+0x21/0x30
 kasan_save_free_info+0x2a/0x50
 __kasan_slab_free+0x106/0x180
 __kmem_cache_free+0xbc/0x320
 clk_unregister+0x6de/0x8d0
 macb_remove+0x73/0xc0 [macb_pci]
 pci_device_remove+0xae/0x210
 device_remove+0xcb/0x180
 device_release_driver_internal+0x529/0x770
 driver_detach+0xd4/0x1a0
 bus_remove_driver+0x135/0x260
 driver_unregister+0x72/0xb0
 pci_unregister_driver+0x26/0x220
 __do_sys_delete_module+0x32e/0x550
 do_syscall_64+0x35/0x80
 entry_SYSCALL_64_after_hwframe+0x6e/0xd8
Fixes d82d5303c4c5 ("net: macb: fix use after free on rmmod")
---
 drivers/net/ethernet/cadence/macb_pci.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/net/ethernet/cadence/macb_pci.c +4 -2
index 3593b310c325d..9be050c5d83e0 100644
@@ -110,10 +110,12 @@ static void macb_remove(struct pci_dev *pdev)
 {
 	struct platform_device *plat_dev = pci_get_drvdata(pdev);
 	struct macb_platform_data *plat_data = dev_get_platdata(&plat_dev->dev);
+	struct clk *pclk = plat_data->pclk;
+	struct clk *hclk = plat_data->hclk;
 
-	clk_unregister(plat_data->pclk);
-	clk_unregister(plat_data->hclk);
 	platform_device_unregister(plat_dev);
+	clk_unregister(pclk);
+	clk_unregister(hclk);
 }
 
 static const struct pci_device_id dev_id_table[] = {
Signature · 1 line
2.53.0
author maintainer
#371 lore raw

PATCH 347/491 net: macb: properly unregister fixed rate clocks

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Fedor Pchelkin <[email protected]>

[ Upstream commit f0f367a4f459cc8118aadc43c6bba53c60d93f8d ]

The additional resources allocated with clk_register_fixed_rate() need to be released with clk_unregister_fixed_rate(), otherwise they are lost.

Fixes 83a77e9ec415 ("net: macb: Added PCI wrapper for Platform Driver.")
---
 drivers/net/ethernet/cadence/macb_pci.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/net/ethernet/cadence/macb_pci.c +4 -4
index 9be050c5d83e0..e6c17e0af1ccf 100644
@@ -97,10 +97,10 @@ static int macb_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	return 0;
 
 err_plat_dev_register:
-	clk_unregister(plat_data.hclk);
+	clk_unregister_fixed_rate(plat_data.hclk);
 
 err_hclk_register:
-	clk_unregister(plat_data.pclk);
+	clk_unregister_fixed_rate(plat_data.pclk);
 
 err_pclk_register:
 	return err;
@@ -114,8 +114,8 @@ static void macb_remove(struct pci_dev *pdev)
 	struct clk *hclk = plat_data->hclk;
 
 	platform_device_unregister(plat_dev);
-	clk_unregister(pclk);
-	clk_unregister(hclk);
+	clk_unregister_fixed_rate(pclk);
+	clk_unregister_fixed_rate(hclk);
 }
 
 static const struct pci_device_id dev_id_table[] = {
Signature · 1 line
2.53.0
author maintainer
#372 lore raw

PATCH 348/491 net/mlx5: Avoid "No data available" when FW version queries fail

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Saeed Mahameed <[email protected]>

[ Upstream commit 10dc35f6a443d488f219d1a1e3fb8f8dac422070 ]

Avoid printing the misleading "kernel answers: No data available" devlink output when querying firmware or pending firmware version fails (e.g. MLX5 fw state errors / flash failures).

FW can fail on loading the pending flash image and get its version due to various reasons, examples:

mlxfw: Firmware flash failed: key not applicable, err (7)
mlx5_fw_image_pending: can't read pending fw version while fw state is 1

and the resulting:

$ devlink dev info
kernel answers: No data available

Instead, just report 0 or 0xfff.. versions in case of failure to indicate a problem, and let other information be shown.

after the fix:

$ devlink dev info
pci/0000:00:06.0:
  driver mlx5_core
  serial_number xxx...
  board.serial_number MT2225300179
  versions:
      fixed:
        fw.psid MT_0000000436
      running:
        fw.version 22.41.0188
        fw 22.41.0188
      stored:
        fw.version 255.255.65535
        fw 255.255.65535
---
 .../net/ethernet/mellanox/mlx5/core/devlink.c |  4 +-
 drivers/net/ethernet/mellanox/mlx5/core/fw.c  | 53 ++++++++++++-------
 .../ethernet/mellanox/mlx5/core/mlx5_core.h   |  4 +-
 3 files changed, 37 insertions(+), 24 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/devlink.c +1 -3
index 060561f633114..62ac22d82e1a9 100644
@@ -60,9 +60,7 @@ mlx5_devlink_info_get(struct devlink *devlink, struct devlink_info_req *req,
 	if (err)
 		return err;
 
-	err = mlx5_fw_version_query(dev, &running_fw, &stored_fw);
-	if (err)
-		return err;
+	mlx5_fw_version_query(dev, &running_fw, &stored_fw);
 
 	snprintf(version_str, sizeof(version_str), "%d.%d.%04d",
 		 mlx5_fw_ver_major(running_fw), mlx5_fw_ver_minor(running_fw),
drivers/net/ethernet/mellanox/mlx5/core/fw.c +34 -19

82 lines not on this page · raw message

drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +2 -2
index b285f1515e4e8..d3279536c902c 100644
@@ -209,8 +209,8 @@ void mlx5_dm_cleanup(struct mlx5_core_dev *dev);
 
 int mlx5_firmware_flash(struct mlx5_core_dev *dev, const struct firmware *fw,
 			struct netlink_ext_ack *extack);
-int mlx5_fw_version_query(struct mlx5_core_dev *dev,
-			  u32 *running_ver, u32 *stored_ver);
+void mlx5_fw_version_query(struct mlx5_core_dev *dev, u32 *running_ver,
+			   u32 *stored_ver);
 
 void mlx5e_init(void);
 void mlx5e_cleanup(void);
Signature · 1 line
2.53.0
author maintainer
#373 lore raw

PATCH 349/491 net/x25: Fix potential double free of skb

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Martin Schiller <[email protected]>

[ Upstream commit d10a26aa4d072320530e6968ef945c8c575edf61 ]

When alloc_skb fails in x25_queue_rx_frame it calls kfree_skb(skb) at line 48 and returns 1 (error).
This error propagates back through the call chain:

x25_queue_rx_frame returns 1

    |
    v
x25_state3_machine receives the return value 1 and takes the else
branch at line 278, setting queued=0 and returning 0
    |
    v
x25_process_rx_frame returns queued=0
    |
    v

x25_backlog_rcv at line 452 sees queued=0 and calls kfree_skb(skb) again

This would free the same skb twice. Looking at x25_backlog_rcv:

net/x25/x25_in.c:x25_backlog_rcv() {
    ...
    queued = x25_process_rx_frame(sk, skb);
    ...
    if (!queued)
        kfree_skb(skb);
}
---
 net/x25/x25_in.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
net/x25/x25_in.c +2 -3
index e1c4197af468e..77ad186507f64 100644
@@ -44,10 +44,9 @@ static int x25_queue_rx_frame(struct sock *sk, struct sk_buff *skb, int more)
 	if (!more && x25->fraglen > 0) {	/* End of fragment */
 		int len = x25->fraglen + skb->len;
 
-		if ((skbn = alloc_skb(len, GFP_ATOMIC)) == NULL){
-			kfree_skb(skb);
+		skbn = alloc_skb(len, GFP_ATOMIC);
+		if (!skbn)
 			return 1;
-		}
 
 		skb_queue_tail(&x25->fragment_queue, skb);
 
Signature · 1 line
2.53.0
author maintainer
#374 lore raw

PATCH 350/491 net/x25: Fix overflow when accumulating packets

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Martin Schiller <[email protected]>

[ Upstream commit a1822cb524e89b4cd2cf0b82e484a2335496a6d9 ]

Add a check to ensure that x25_sock.fraglen does not overflow.

The fraglen also needs to be resetted when purging fragment_queue in x25_clear_queues().

---
 net/x25/x25_in.c   | 4 ++++
 net/x25/x25_subr.c | 1 +
 2 files changed, 5 insertions(+)
net/x25/x25_in.c +4 -0
index 77ad186507f64..956e056803079 100644
@@ -34,6 +34,10 @@ static int x25_queue_rx_frame(struct sock *sk, struct sk_buff *skb, int more)
 	struct sk_buff *skbo, *skbn = skb;
 	struct x25_sock *x25 = x25_sk(sk);
 
+	/* make sure we don't overflow */
+	if (x25->fraglen + skb->len > USHRT_MAX)
+		return 1;
+
 	if (more) {
 		x25->fraglen += skb->len;
 		skb_queue_tail(&x25->fragment_queue, skb);
net/x25/x25_subr.c +1 -0
index 0285aaa1e93c1..159708d9ad20c 100644
@@ -40,6 +40,7 @@ void x25_clear_queues(struct sock *sk)
 	skb_queue_purge(&x25->interrupt_in_queue);
 	skb_queue_purge(&x25->interrupt_out_queue);
 	skb_queue_purge(&x25->fragment_queue);
+	x25->fraglen = 0;
 }
 
 
Signature · 1 line
2.53.0
author maintainer
#375 lore raw

PATCH 351/491 net/sched: cls_fw: fix NULL pointer dereference on shared blocks

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Xiang Mei <[email protected]>

[ Upstream commit faeea8bbf6e958bf3c00cb08263109661975987c ]

The old-method path in fw_classify() calls tcf_block_q() and dereferences q->handle. Shared blocks leave block->q NULL, causing a NULL deref when an empty cls_fw filter is attached to a shared block and a packet with a nonzero major skb mark is classified.

Reject the configuration in fw_change() when the old method (no TCA_OPTIONS) is used on a shared block, since fw_classify()'s old-method path needs block->q which is NULL for shared blocks.

The fixed null-ptr-deref calling stack:

 KASAN: null-ptr-deref in range [0x0000000000000038-0x000000000000003f]
 RIP: 0010:fw_classify (net/sched/cls_fw.c:81)
 Call Trace:
  tcf_classify (./include/net/tc_wrapper.h:197 net/sched/cls_api.c:1764 net/sched/cls_api.c:1860)
  tc_run (net/core/dev.c:4401)
  __dev_queue_xmit (net/core/dev.c:4535 net/core/dev.c:4790)
Fixes 1abf272022cf ("net: sched: tcindex, fw, flow: use tcf_block_q helper to get struct Qdisc")
---
 net/sched/cls_fw.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)
net/sched/cls_fw.c +12 -2
index 08c41f1976c47..23cf4f7111174 100644
@@ -246,8 +246,18 @@ static int fw_change(struct net *net, struct sk_buff *in_skb,
 	struct nlattr *tb[TCA_FW_MAX + 1];
 	int err;
 
-	if (!opt)
-		return handle ? -EINVAL : 0; /* Succeed if it is old method. */
+	if (!opt) {
+		if (handle)
+			return -EINVAL;
+
+		if (tcf_block_shared(tp->chain->block)) {
+			NL_SET_ERR_MSG(extack,
+				       "Must specify mark when attaching fw filter to block");
+			return -EINVAL;
+		}
+
+		return 0; /* Succeed if it is old method. */
+	}
 
 	err = nla_parse_nested_deprecated(tb, TCA_FW_MAX, opt, fw_policy,
 					  NULL);
Signature · 1 line
2.53.0
author maintainer
#376 lore raw

PATCH 352/491 net/sched: cls_flow: fix NULL pointer dereference on shared blocks

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Xiang Mei <[email protected]>

[ Upstream commit 1a280dd4bd1d616a01d6ffe0de284c907b555504 ]

flow_change() calls tcf_block_q() and dereferences q->handle to derive a default baseclass. Shared blocks leave block->q NULL, causing a NULL deref when a flow filter without a fully qualified baseclass is created on a shared block.

Check tcf_block_shared() before accessing block->q and return -EINVAL for shared blocks. This avoids the null-deref shown below:

=======================================================================

KASAN: null-ptr-deref in range [0x0000000000000038-0x000000000000003f]
RIP: 0010:flow_change (net/sched/cls_flow.c:508)
Call Trace:
 tc_new_tfilter (net/sched/cls_api.c:2432)
 rtnetlink_rcv_msg (net/core/rtnetlink.c:6980)
 [...]
=======================================================================
Fixes 1abf272022cf ("net: sched: tcindex, fw, flow: use tcf_block_q helper to get struct Qdisc")
---
 net/sched/cls_flow.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)
net/sched/cls_flow.c +9 -1
index 117c7b038591e..7918ecdcfe696 100644
@@ -501,8 +501,16 @@ static int flow_change(struct net *net, struct sk_buff *in_skb,
 		}
 
 		if (TC_H_MAJ(baseclass) == 0) {
-			struct Qdisc *q = tcf_block_q(tp->chain->block);
+			struct tcf_block *block = tp->chain->block;
+			struct Qdisc *q;
 
+			if (tcf_block_shared(block)) {
+				NL_SET_ERR_MSG(extack,
+					       "Must specify baseclass when attaching flow filter to block");
+				goto err2;
+			}
+
+			q = tcf_block_q(block);
 			baseclass = TC_H_MAKE(q->handle, baseclass);
 		}
 		if (TC_H_MIN(baseclass) == 0)
Signature · 1 line
2.53.0
author maintainer
#377 lore raw

PATCH 353/491 ipv6: avoid overflows in ip6_datagram_send_ctl()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 4e453375561fc60820e6b9d8ebeb6b3ee177d42e ]

Yiming Qian reported :
<quote>
I believe I found a locally triggerable kernel bug in the IPv6 sendmsg ancillary-data path that can panic the kernel via skb_under_panic() (local DoS).

The core issue is a mismatch between:

- a 16-bit length accumulator (struct ipv6_txoptions::opt_flen, type __u16) and
- a pointer to the *last* provided destination-options header (opt->dst1opt)

when multiple IPV6_DSTOPTS control messages (cmsgs) are provided.

- include/net/ipv6.h:
- struct ipv6_txoptions::opt_flen is __u16 (wrap possible).
(lines 291-307, especially 298)
- net/ipv6/datagram.c:ip6_datagram_send_ctl():
- Accepts repeated IPV6_DSTOPTS and accumulates into opt_flen
without rejecting duplicates. (lines 909-933)
- net/ipv6/ip6_output.c:__ip6_append_data():
- Uses opt->opt_flen + opt->opt_nflen to compute header
sizes/headroom decisions. (lines 1448-1466, especially 1463-1465)
- net/ipv6/ip6_output.c:__ip6_make_skb():
- Calls ipv6_push_frag_opts() if opt->opt_flen is non-zero.
(lines 1930-1934)
- net/ipv6/exthdrs.c:ipv6_push_frag_opts() / ipv6_push_exthdr():
- Push size comes from ipv6_optlen(opt->dst1opt) (based on the
pointed-to header). (lines 1179-1185 and 1206-1211)

1. opt_flen is a 16-bit accumulator:

- include/net/ipv6.h:298 defines __u16 opt_flen; /* after fragment hdr */.

2. ip6_datagram_send_ctl() accepts *repeated* IPV6_DSTOPTS cmsgs and increments opt_flen each time:

- In net/ipv6/datagram.c:909-933, for IPV6_DSTOPTS:
- It computes len = ((hdr->hdrlen + 1) << 3);
- It checks CAP_NET_RAW using `ns_capable(net->user_ns,
CAP_NET_RAW)`. (line 922)
- Then it does:
- opt->opt_flen += len; (line 927)
- opt->dst1opt = hdr; (line 928)

There is no duplicate rejection here (unlike the legacy IPV6_2292DSTOPTS path which rejects duplicates at net/ipv6/datagram.c:901-904).

If enough large IPV6_DSTOPTS cmsgs are provided, opt_flen wraps while dst1opt still points to a large (2048-byte) destination-options header.

In the attached PoC (poc.c):

- 32 cmsgs with hdrlen=255 => len = (255+1)*8 = 2048
- 1 cmsg with hdrlen=0 => len = 8
- Total increment: 32*2048 + 8 = 65544, so (__u16)opt_flen == 8
- The last cmsg is 2048 bytes, so dst1opt points to a 2048-byte header.

3. The transmit path sizes headers using the wrapped opt_flen:

- In net/ipv6/ip6_output.c:1463-1465:
- `headersize = sizeof(struct ipv6hdr) + (opt ? opt->opt_flen +
opt->opt_nflen : 0) + ...;`

With wrapped opt_flen, headersize/headroom decisions underestimate what will be pushed later.

4. When building the final skb, the actual push length comes from dst1opt and is not limited by wrapped opt_flen:

- In net/ipv6/ip6_output.c:1930-1934:
- if (opt->opt_flen) proto = ipv6_push_frag_opts(skb, opt, proto);
- In net/ipv6/exthdrs.c:1206-1211, ipv6_push_frag_opts() pushes dst1opt via ipv6_push_exthdr().
- In net/ipv6/exthdrs.c:1179-1184, ipv6_push_exthdr() does:
- skb_push(skb, ipv6_optlen(opt));
- memcpy(h, opt, ipv6_optlen(opt));

With insufficient headroom, skb_push() underflows and triggers skb_under_panic() -> BUG():

- net/core/skbuff.c:2669-2675 (skb_push() calls skb_under_panic())
- net/core/skbuff.c:207-214 (skb_panic() ends in BUG())

- The IPV6_DSTOPTS cmsg path requires CAP_NET_RAW in the target netns user namespace (ns_capable(net->user_ns, CAP_NET_RAW)).
- Root (or any task with CAP_NET_RAW) can trigger this without user namespaces.
- An unprivileged uid=1000 user can trigger this if unprivileged user namespaces are enabled and it can create a userns+netns to obtain namespaced CAP_NET_RAW (the attached PoC does this).

- Local denial of service: kernel BUG/panic (system crash).
- Reproducible with a small userspace PoC.
</quote>

This patch does not reject duplicated options, as this might break some user applications.

Instead, it makes sure to adjust opt_flen and opt_nflen to correctly reflect the size of the current option headers, preventing the overflows and the potential for panics.

This applies to IPV6_DSTOPTS, IPV6_HOPOPTS, and IPV6_RTHDR.

Specifically:

When a new IPV6_DSTOPTS is processed, the length of the old opt->dst1opt is subtracted from opt->opt_flen before adding the new length.

When a new IPV6_HOPOPTS is processed, the length of the old opt->dst0opt is subtracted from opt->opt_nflen.

When a new Routing Header (IPV6_RTHDR or IPV6_2292RTHDR) is processed, the length of the old opt->srcrt is subtracted from opt->opt_nflen.

In the special case within IPV6_2292RTHDR handling where dst1opt is moved to dst0opt, the length of the old opt->dst0opt is subtracted from opt->opt_nflen before the new one is added.

---
 net/ipv6/datagram.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
net/ipv6/datagram.c +10 -0
index a30ff5d6808aa..d8af31805133f 100644
@@ -756,6 +756,7 @@ int ip6_datagram_send_ctl(struct net *net, struct sock *sk,
 {
 	struct in6_pktinfo *src_info;
 	struct cmsghdr *cmsg;
+	struct ipv6_rt_hdr *orthdr;
 	struct ipv6_rt_hdr *rthdr;
 	struct ipv6_opt_hdr *hdr;
 	struct ipv6_txoptions *opt = ipc6->opt;
@@ -917,9 +918,13 @@ int ip6_datagram_send_ctl(struct net *net, struct sock *sk,
 				goto exit_f;
 			}
 			if (cmsg->cmsg_type == IPV6_DSTOPTS) {
+				if (opt->dst1opt)
+					opt->opt_flen -= ipv6_optlen(opt->dst1opt);
 				opt->opt_flen += len;
 				opt->dst1opt = hdr;
 			} else {
+				if (opt->dst0opt)
+					opt->opt_nflen -= ipv6_optlen(opt->dst0opt);
 				opt->opt_nflen += len;
 				opt->dst0opt = hdr;
 			}
@@ -962,12 +967,17 @@ int ip6_datagram_send_ctl(struct net *net, struct sock *sk,
 				goto exit_f;
 			}
 
+			orthdr = opt->srcrt;
+			if (orthdr)
+				opt->opt_nflen -= ((orthdr->hdrlen + 1) << 3);
 			opt->opt_nflen += len;
 			opt->srcrt = rthdr;
 
 			if (cmsg->cmsg_type == IPV6_2292RTHDR && opt->dst1opt) {
 				int dsthdrlen = ((opt->dst1opt->hdrlen+1)<<3);
 
+				if (opt->dst0opt)
+					opt->opt_nflen -= ipv6_optlen(opt->dst0opt);
 				opt->opt_nflen += dsthdrlen;
 				opt->dst0opt = opt->dst1opt;
 				opt->dst1opt = NULL;
Signature · 1 line
2.53.0
author maintainer
#378 lore raw

PATCH 354/491 efi/mokvar-table: Avoid repeated map/unmap of the same page

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

[ Upstream commit e3cf2d91d0583cae70aeb512da87e3ade25ea912 ]

Tweak the logic that traverses the MOKVAR UEFI configuration table to only unmap the entry header and map the next one if they don't live in the same physical page.

---
 drivers/firmware/efi/mokvar-table.c | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)
drivers/firmware/efi/mokvar-table.c +15 -3
index 39a49adf007a5..69ac876ca809d 100644
@@ -99,13 +99,13 @@ static struct kobject *mokvar_kobj;
  */
 void __init efi_mokvar_table_init(void)
 {
+	struct efi_mokvar_table_entry __aligned(1) *mokvar_entry, *next_entry;
 	efi_memory_desc_t md;
 	void *va = NULL;
 	unsigned long cur_offset = 0;
 	unsigned long offset_limit;
 	unsigned long map_size_needed = 0;
 	unsigned long size;
-	struct efi_mokvar_table_entry *mokvar_entry;
 	int err;
 
 	if (!efi_enabled(EFI_MEMMAP))
@@ -142,7 +142,7 @@ void __init efi_mokvar_table_init(void)
 			return;
 		}
 		mokvar_entry = va;
-
+next:
 		/* Check for last sentinel entry */
 		if (mokvar_entry->name[0] == '\0') {
 			if (mokvar_entry->data_size != 0)
@@ -156,7 +156,19 @@ void __init efi_mokvar_table_init(void)
 		mokvar_entry->name[sizeof(mokvar_entry->name) - 1] = '\0';
 
 		/* Advance to the next entry */
-		cur_offset += sizeof(*mokvar_entry) + mokvar_entry->data_size;
+		size = sizeof(*mokvar_entry) + mokvar_entry->data_size;
+		cur_offset += size;
+
+		/*
+		 * Don't bother remapping if the current entry header and the
+		 * next one end on the same page.
+		 */
+		next_entry = (void *)((unsigned long)mokvar_entry + size);
+		if (((((unsigned long)(mokvar_entry + 1) - 1) ^
+		      ((unsigned long)(next_entry + 1) - 1)) & PAGE_MASK) == 0) {
+			mokvar_entry = next_entry;
+			goto next;
+		}
 	}
 
 	if (va)
Signature · 1 line
2.53.0
author maintainer
#379 lore raw

PATCH 355/491 Revert "drm/vmwgfx: Add seqno waiter for sync_files"

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

This reverts commit 7db6c88bb52f3b7525a06110cfd208990c49f8b4.

---
 drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 26 -------------------------
 1 file changed, 26 deletions(-)
drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +0 -26
index 361f96d09374d..daea547704ddc 100644
@@ -4039,23 +4039,6 @@ static int vmw_execbuf_tie_context(struct vmw_private *dev_priv,
 	return 0;
 }
 
-/*
- * DMA fence callback to remove a seqno_waiter
- */
-struct seqno_waiter_rm_context {
-	struct dma_fence_cb base;
-	struct vmw_private *dev_priv;
-};
-
-static void seqno_waiter_rm_cb(struct dma_fence *f, struct dma_fence_cb *cb)
-{
-	struct seqno_waiter_rm_context *ctx =
-		container_of(cb, struct seqno_waiter_rm_context, base);
-
-	vmw_seqno_waiter_remove(ctx->dev_priv);
-	kfree(ctx);
-}
-
 int vmw_execbuf_process(struct drm_file *file_priv,
 			struct vmw_private *dev_priv,
 			void __user *user_commands, void *kernel_commands,
@@ -4249,15 +4232,6 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 		} else {
 			/* Link the fence with the FD created earlier */
 			fd_install(out_fence_fd, sync_file->file);
-			struct seqno_waiter_rm_context *ctx =
-				kmalloc(sizeof(*ctx), GFP_KERNEL);
-			ctx->dev_priv = dev_priv;
-			vmw_seqno_waiter_add(dev_priv);
-			if (dma_fence_add_callback(&fence->base, &ctx->base,
-						   seqno_waiter_rm_cb) < 0) {
-				vmw_seqno_waiter_remove(dev_priv);
-				kfree(ctx);
-			}
 		}
 	}
 
Signature · 1 line
2.53.0
author maintainer
#380 lore raw

PATCH 356/491 drm/vmwgfx: Add seqno waiter for sync_files

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ian Forbes <[email protected]>

[ Upstream commit 0039a3b35b10d9c15d3d26320532ab56cc566750 ]

Because sync_files are passive waiters they do not participate in the processing of fences like the traditional vmw_fence_wait IOCTL. If userspace exclusively uses sync_files for synchronization then nothing in the kernel actually processes fence updates as interrupts for fences are masked and ignored if the kernel does not indicate to the SVGA device that there are active waiters.

This oversight results in a bug where the entire GUI can freeze waiting on a sync_file that will never be signalled as we've masked the interrupts to signal its completion. This bug is incredibly racy as any process which interacts with the fencing code via the 3D stack can process the stuck fences on behalf of the stuck process causing it to run again. Even a simple app like eglinfo is enough to resume the stuck process. Usually this bug is seen at a login screen like GDM because there are no other 3D apps running.

By adding a seqno waiter we re-enable interrupt based processing of the dma_fences associated with the sync_file which is signalled as part of a dma_fence_callback.

This has likely been broken since it was initially added to the kernel in 2017 but has gone unnoticed until mutter recently started using sync_files heavily over the course of 2024 as part of their explicit sync support.

Fixes c906965dee22 ("drm/vmwgfx: Add export fence to file descriptor support")
---
 drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 26 +++++++++++++++++++++++++
 1 file changed, 26 insertions(+)
drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +26 -0
index daea547704ddc..9bec26fda14cb 100644
@@ -4039,6 +4039,23 @@ static int vmw_execbuf_tie_context(struct vmw_private *dev_priv,
 	return 0;
 }
 
+/*
+ * DMA fence callback to remove a seqno_waiter
+ */
+struct seqno_waiter_rm_context {
+	struct dma_fence_cb base;
+	struct vmw_private *dev_priv;
+};
+
+static void seqno_waiter_rm_cb(struct dma_fence *f, struct dma_fence_cb *cb)
+{
+	struct seqno_waiter_rm_context *ctx =
+		container_of(cb, struct seqno_waiter_rm_context, base);
+
+	vmw_seqno_waiter_remove(ctx->dev_priv);
+	kfree(ctx);
+}
+
 int vmw_execbuf_process(struct drm_file *file_priv,
 			struct vmw_private *dev_priv,
 			void __user *user_commands, void *kernel_commands,
@@ -4230,8 +4247,17 @@ int vmw_execbuf_process(struct drm_file *file_priv,
 			fput(sync_file->file);
 			put_unused_fd(out_fence_fd);
 		} else {
+			struct seqno_waiter_rm_context *ctx;
 			/* Link the fence with the FD created earlier */
 			fd_install(out_fence_fd, sync_file->file);
+			ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
+			ctx->dev_priv = dev_priv;
+			vmw_seqno_waiter_add(dev_priv);
+			if (dma_fence_add_callback(&fence->base, &ctx->base,
+						   seqno_waiter_rm_cb) < 0) {
+				vmw_seqno_waiter_remove(dev_priv);
+				kfree(ctx);
+			}
 		}
 	}
 
Signature · 1 line
2.53.0
author maintainer
#381 lore raw

PATCH 357/491 Revert "scsi: core: Wake up the error handler when final completions race against each other"

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

This reverts commit cc872e35c0df80062abc71268d690a2f749e542e.

---
 drivers/scsi/scsi_error.c | 11 +----------
 drivers/scsi/scsi_lib.c   |  8 --------
 2 files changed, 1 insertion(+), 18 deletions(-)
drivers/scsi/scsi_error.c +1 -10
index 4e9114f069832..ffc6f3031e82b 100644
@@ -241,20 +241,11 @@ static void scsi_eh_inc_host_failed(struct rcu_head *head)
 {
 	struct scsi_cmnd *scmd = container_of(head, typeof(*scmd), rcu);
 	struct Scsi_Host *shost = scmd->device->host;
-	unsigned int busy;
+	unsigned int busy = scsi_host_busy(shost);
 	unsigned long flags;
 
 	spin_lock_irqsave(shost->host_lock, flags);
 	shost->host_failed++;
-	spin_unlock_irqrestore(shost->host_lock, flags);
-	/*
-	 * The counting of busy requests needs to occur after adding to
-	 * host_failed or after the lock acquire for adding to host_failed
-	 * to prevent a race with host unbusy and missing an eh wakeup.
-	 */
-	busy = scsi_host_busy(shost);
-
-	spin_lock_irqsave(shost->host_lock, flags);
 	scsi_eh_wakeup(shost, busy);
 	spin_unlock_irqrestore(shost->host_lock, flags);
 }
drivers/scsi/scsi_lib.c +0 -8
index 8d570632982f3..fb48d47e9183e 100644
@@ -310,14 +310,6 @@ static void scsi_dec_host_busy(struct Scsi_Host *shost, struct scsi_cmnd *cmd)
 	rcu_read_lock();
 	__clear_bit(SCMD_STATE_INFLIGHT, &cmd->state);
 	if (unlikely(scsi_host_in_recovery(shost))) {
-		/*
-		 * Ensure the clear of SCMD_STATE_INFLIGHT is visible to
-		 * other CPUs before counting busy requests. Otherwise,
-		 * reordering can cause CPUs to race and miss an eh wakeup
-		 * when no CPU sees all busy requests as done or timed out.
-		 */
-		smp_mb();
-
 		unsigned int busy = scsi_host_busy(shost);
 
 		spin_lock_irqsave(shost->host_lock, flags);
Signature · 1 line
2.53.0
author maintainer
#382 lore raw

PATCH 358/491 scsi: core: Wake up the error handler when final completions race against each other

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Jeffery <[email protected]>

[ Upstream commit fe2f8ad6f0999db3b318359a01ee0108c703a8c3 ]

The fragile ordering between marking commands completed or failed so that the error handler only wakes when the last running command completes or times out has race conditions. These race conditions can cause the SCSI layer to fail to wake the error handler, leaving I/O through the SCSI host stuck as the error state cannot advance.

First, there is an memory ordering issue within scsi_dec_host_busy(). The write which clears SCMD_STATE_INFLIGHT may be reordered with reads counting in scsi_host_busy(). While the local CPU will see its own write, reordering can allow other CPUs in scsi_dec_host_busy() or scsi_eh_inc_host_failed() to see a raised busy count, causing no CPU to see a host busy equal to the host_failed count.

This race condition can be prevented with a memory barrier on the error path to force the write to be visible before counting host busy commands.

Second, there is a general ordering issue with scsi_eh_inc_host_failed(). By counting busy commands before incrementing host_failed, it can race with a final command in scsi_dec_host_busy(), such that scsi_dec_host_busy() does not see host_failed incremented but scsi_eh_inc_host_failed() counts busy commands before SCMD_STATE_INFLIGHT is cleared by scsi_dec_host_busy(), resulting in neither waking the error handler task.

This needs the call to scsi_host_busy() to be moved after host_failed is incremented to close the race condition.

Fixes 6eb045e092ef ("scsi: core: avoid host-wide host_busy counter for scsi_mq")
---
 drivers/scsi/scsi_error.c | 11 ++++++++++-
 drivers/scsi/scsi_lib.c   | 11 ++++++++++-
 2 files changed, 20 insertions(+), 2 deletions(-)
drivers/scsi/scsi_error.c +10 -1
index ffc6f3031e82b..4e9114f069832 100644
@@ -241,11 +241,20 @@ static void scsi_eh_inc_host_failed(struct rcu_head *head)
 {
 	struct scsi_cmnd *scmd = container_of(head, typeof(*scmd), rcu);
 	struct Scsi_Host *shost = scmd->device->host;
-	unsigned int busy = scsi_host_busy(shost);
+	unsigned int busy;
 	unsigned long flags;
 
 	spin_lock_irqsave(shost->host_lock, flags);
 	shost->host_failed++;
+	spin_unlock_irqrestore(shost->host_lock, flags);
+	/*
+	 * The counting of busy requests needs to occur after adding to
+	 * host_failed or after the lock acquire for adding to host_failed
+	 * to prevent a race with host unbusy and missing an eh wakeup.
+	 */
+	busy = scsi_host_busy(shost);
+
+	spin_lock_irqsave(shost->host_lock, flags);
 	scsi_eh_wakeup(shost, busy);
 	spin_unlock_irqrestore(shost->host_lock, flags);
 }
drivers/scsi/scsi_lib.c +10 -1
index fb48d47e9183e..e69d1c0ea4507 100644
@@ -310,7 +310,16 @@ static void scsi_dec_host_busy(struct Scsi_Host *shost, struct scsi_cmnd *cmd)
 	rcu_read_lock();
 	__clear_bit(SCMD_STATE_INFLIGHT, &cmd->state);
 	if (unlikely(scsi_host_in_recovery(shost))) {
-		unsigned int busy = scsi_host_busy(shost);
+		unsigned int busy;
+		/*
+		 * Ensure the clear of SCMD_STATE_INFLIGHT is visible to
+		 * other CPUs before counting busy requests. Otherwise,
+		 * reordering can cause CPUs to race and miss an eh wakeup
+		 * when no CPU sees all busy requests as done or timed out.
+		 */
+		smp_mb();
+
+		busy = scsi_host_busy(shost);
 
 		spin_lock_irqsave(shost->host_lock, flags);
 		if (shost->host_failed || shost->host_eh_scheduled)
Signature · 1 line
2.53.0
author maintainer
#383 lore raw

PATCH 359/491 Revert "media: dvb-frontends: w7090p: fix null-ptr-deref in w7090p_tuner_write_serpar and w7090p_tuner_read_serpar"

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

This reverts commit b3d77a3fc71c084575d3df4ec6544b3fb6ce587d.

---
 drivers/media/dvb-frontends/dib7000p.c | 4 ----
 1 file changed, 4 deletions(-)
drivers/media/dvb-frontends/dib7000p.c +0 -4
index 08b3ac8ff1083..a4d060fb1babd 100644
@@ -2198,8 +2198,6 @@ static int w7090p_tuner_write_serpar(struct i2c_adapter *i2c_adap, struct i2c_ms
 	struct dib7000p_state *state = i2c_get_adapdata(i2c_adap);
 	u8 n_overflow = 1;
 	u16 i = 1000;
-	if (msg[0].len < 3)
-		return -EOPNOTSUPP;
 	u16 serpar_num = msg[0].buf[0];
 
 	while (n_overflow == 1 && i) {
@@ -2219,8 +2217,6 @@ static int w7090p_tuner_read_serpar(struct i2c_adapter *i2c_adap, struct i2c_msg
 	struct dib7000p_state *state = i2c_get_adapdata(i2c_adap);
 	u8 n_overflow = 1, n_empty = 1;
 	u16 i = 1000;
-	if (msg[0].len < 1 || msg[1].len < 2)
-		return -EOPNOTSUPP;
 	u16 serpar_num = msg[0].buf[0];
 	u16 read_word;
 
Signature · 1 line
2.53.0
author maintainer
#384 lore raw

PATCH 360/491 media: dvb-frontends: w7090p: fix null-ptr-deref in w7090p_tuner_write_serpar and w7090p_tuner_read_serpar

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Guo <[email protected]>

[ Upstream commit ed0234c8458b3149f15e496b48a1c9874dd24a1b ]

In w7090p_tuner_write_serpar, msg is controlled by user. When msg[0].buf is null and msg[0].len is zero, former checks on msg[0].buf would be passed. If accessing msg[0].buf[2] without sanity check, null pointer deref would happen. We add check on msg[0].len to prevent crash.

Similar commit: commit 0ed554fd769a ("media: dvb-usb: az6027: fix null-ptr-deref in az6027_i2c_xfer()")

---
 drivers/media/dvb-frontends/dib7000p.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)
drivers/media/dvb-frontends/dib7000p.c +10 -2
index a4d060fb1babd..18b48a3984291 100644
@@ -2198,7 +2198,11 @@ static int w7090p_tuner_write_serpar(struct i2c_adapter *i2c_adap, struct i2c_ms
 	struct dib7000p_state *state = i2c_get_adapdata(i2c_adap);
 	u8 n_overflow = 1;
 	u16 i = 1000;
-	u16 serpar_num = msg[0].buf[0];
+	u16 serpar_num;
+
+	if (msg[0].len < 3)
+		return -EOPNOTSUPP;
+	serpar_num = msg[0].buf[0];
 
 	while (n_overflow == 1 && i) {
 		n_overflow = (dib7000p_read_word(state, 1984) >> 1) & 0x1;
@@ -2217,9 +2221,13 @@ static int w7090p_tuner_read_serpar(struct i2c_adapter *i2c_adap, struct i2c_msg
 	struct dib7000p_state *state = i2c_get_adapdata(i2c_adap);
 	u8 n_overflow = 1, n_empty = 1;
 	u16 i = 1000;
-	u16 serpar_num = msg[0].buf[0];
+	u16 serpar_num;
 	u16 read_word;
 
+	if (msg[0].len < 1 || msg[1].len < 2)
+		return -EOPNOTSUPP;
+	serpar_num = msg[0].buf[0];
+
 	while (n_overflow == 1 && i) {
 		n_overflow = (dib7000p_read_word(state, 1984) >> 1) & 0x1;
 		i--;
Signature · 1 line
2.53.0
author maintainer
#385 lore raw

PATCH 361/491 hwmon: (pxe1610) Check return value of page-select write in probe

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sanman Pradhan <[email protected]>

[ Upstream commit ccf70c41e562b29d1c05d1bbf53391785e09c6fb ]

pxe1610_probe() writes PMBUS_PAGE to select page 0 but does not check the return value. If the write fails, subsequent register reads operate on an indeterminate page, leading to silent misconfiguration.

Check the return value and propagate the error using dev_err_probe(), which also handles -EPROBE_DEFER correctly without log spam.

Fixes 344757bac526 ("hwmon: (pmbus) Add Infineon PXE1610 VR driver")
[groeck: Fix "Fixes" SHA]
---
 drivers/hwmon/pmbus/pxe1610.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/hwmon/pmbus/pxe1610.c +4 -1
index 212433eb6cc31..7794e5cf550fc 100644
@@ -104,7 +104,10 @@ static int pxe1610_probe(struct i2c_client *client)
 	 * By default this device doesn't boot to page 0, so set page 0
 	 * to access all pmbus registers.
 	 */
-	i2c_smbus_write_byte_data(client, PMBUS_PAGE, 0);
+	ret = i2c_smbus_write_byte_data(client, PMBUS_PAGE, 0);
+	if (ret < 0)
+		return dev_err_probe(&client->dev, ret,
+				     "Failed to set page 0\n");
 
 	/* Read Manufacturer id */
 	ret = i2c_smbus_read_block_data(client, PMBUS_MFR_ID, buf);
Signature · 1 line
2.53.0
author maintainer
#386 lore raw

PATCH 362/491 hwmon: (occ) Fix missing newline in occ_show_extended()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sanman Pradhan <[email protected]>

[ Upstream commit 09773978879ecf71a7990fe9a28ce4eb92bce645 ]

In occ_show_extended() case 0, when the EXTN_FLAG_SENSOR_ID flag is set, the sysfs_emit format string "%u" is missing the trailing newline that the sysfs ABI expects. The else branch correctly uses "%4phN\n", and all other show functions in this file include the trailing newline.

Add the missing "\n" for consistency and correct sysfs output.

Fixes c10e753d43eb ("hwmon (occ): Add sensor types and versions")
---
 drivers/hwmon/occ/common.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/hwmon/occ/common.c +1 -1
index 1a6c8ebc32e16..2b088418259b0 100644
@@ -727,7 +727,7 @@ static ssize_t occ_show_extended(struct device *dev,
 	switch (sattr->nr) {
 	case 0:
 		if (extn->flags & EXTN_FLAG_SENSOR_ID) {
-			rc = sysfs_emit(buf, "%u",
+			rc = sysfs_emit(buf, "%u\n",
 					get_unaligned_be32(&extn->sensor_id));
 		} else {
 			rc = sysfs_emit(buf, "%02x%02x%02x%02x\n",
Signature · 1 line
2.53.0
author maintainer
#387 lore raw

PATCH 363/491 riscv: kgdb: fix several debug register assignment bugs

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paul Walmsley <[email protected]>

[ Upstream commit 834911eb8eef2501485d819b4eabebadc25c3497 ]

Fix several bugs in the RISC-V kgdb implementation:

- The element of dbg_reg_def[] that is supposed to pertain to the S1 register embeds instead the struct pt_regs offset of the A1 register. Fix this to use the S1 register offset in struct pt_regs.

- The sleeping_thread_to_gdb_regs() function copies the value of the S10 register into the gdb_regs[] array element meant for the S9 register, and copies the value of the S11 register into the array element meant for the S10 register. It also neglects to copy the value of the S11 register. Fix all of these issues.

Fixes fe89bd2be8667 ("riscv: Add KGDB support")
Cc Vincent Chen <[email protected]>
---
 arch/riscv/kernel/kgdb.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
arch/riscv/kernel/kgdb.c +4 -3
index 1d83b36967212..eb737c7a563b9 100644
@@ -194,7 +194,7 @@ struct dbg_reg_def_t dbg_reg_def[DBG_MAX_REG_NUM] = {
 	{DBG_REG_T1, GDB_SIZEOF_REG, offsetof(struct pt_regs, t1)},
 	{DBG_REG_T2, GDB_SIZEOF_REG, offsetof(struct pt_regs, t2)},
 	{DBG_REG_FP, GDB_SIZEOF_REG, offsetof(struct pt_regs, s0)},
-	{DBG_REG_S1, GDB_SIZEOF_REG, offsetof(struct pt_regs, a1)},
+	{DBG_REG_S1, GDB_SIZEOF_REG, offsetof(struct pt_regs, s1)},
 	{DBG_REG_A0, GDB_SIZEOF_REG, offsetof(struct pt_regs, a0)},
 	{DBG_REG_A1, GDB_SIZEOF_REG, offsetof(struct pt_regs, a1)},
 	{DBG_REG_A2, GDB_SIZEOF_REG, offsetof(struct pt_regs, a2)},
@@ -263,8 +263,9 @@ sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *task)
 	gdb_regs[DBG_REG_S6_OFF] = task->thread.s[6];
 	gdb_regs[DBG_REG_S7_OFF] = task->thread.s[7];
 	gdb_regs[DBG_REG_S8_OFF] = task->thread.s[8];
-	gdb_regs[DBG_REG_S9_OFF] = task->thread.s[10];
-	gdb_regs[DBG_REG_S10_OFF] = task->thread.s[11];
+	gdb_regs[DBG_REG_S9_OFF] = task->thread.s[9];
+	gdb_regs[DBG_REG_S10_OFF] = task->thread.s[10];
+	gdb_regs[DBG_REG_S11_OFF] = task->thread.s[11];
 	gdb_regs[DBG_REG_EPC_OFF] = task->thread.ra;
 }
 
Signature · 1 line
2.53.0
author maintainer
#388 lore raw

PATCH 364/491 drm/ioc32: stop speculation on the drm_compat_ioctl path

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Greg Kroah-Hartman <[email protected]>

commit f8995c2df519f382525ca4bc90553ad2ec611067 upstream.

The drm compat ioctl path takes a user controlled pointer, and then dereferences it into a table of function pointers, the signature method of spectre problems. Fix this up by calling array_index_nospec() on the index to the function pointer list.

Fixes 505b5240329b ("drm/ioctl: Fix Spectre v1 vulnerabilities")
Cc Maarten Lankhorst <[email protected]>
Cc Maxime Ripard <[email protected]>
Cc Thomas Zimmermann <[email protected]>
Cc David Airlie <[email protected]>
Cc Simona Vetter <[email protected]>
Cc stable <[email protected]>
Assisted-by gkh_clanker_2000
---
 drivers/gpu/drm/drm_ioc32.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/gpu/drm/drm_ioc32.c +2 -0
@@ -28,6 +28,7 @@
  * IN THE SOFTWARE.
  */
 #include <linux/compat.h>
+#include <linux/nospec.h>
 #include <linux/ratelimit.h>
 #include <linux/export.h>
 
@@ -992,6 +993,7 @@ long drm_compat_ioctl(struct file *filp,
 	if (nr >= ARRAY_SIZE(drm_compat_ioctls))
 		return drm_ioctl(filp, cmd, arg);
 
+	nr = array_index_nospec(nr, ARRAY_SIZE(drm_compat_ioctls));
 	fn = drm_compat_ioctls[nr].fn;
 	if (!fn)
 		return drm_ioctl(filp, cmd, arg);
author maintainer
#389 lore raw

PATCH 365/491 wifi: wilc1000: fix u8 overflow in SSID scan buffer size calculation

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yasuaki Torimaru <[email protected]>

commit d049e56b1739101d1c4d81deedb269c52a8dbba0 upstream.

The variable valuesize is declared as u8 but accumulates the total length of all SSIDs to scan. Each SSID contributes up to 33 bytes (IEEE80211_MAX_SSID_LEN + 1), and with WILC_MAX_NUM_PROBED_SSID (10) SSIDs the total can reach 330, which wraps around to 74 when stored in a u8.

This causes kmalloc to allocate only 75 bytes while the subsequent memcpy writes up to 331 bytes into the buffer, resulting in a 256-byte heap buffer overflow.

Widen valuesize from u8 to u32 to accommodate the full range.

---
 drivers/net/wireless/microchip/wilc1000/hif.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/microchip/wilc1000/hif.c +1 -1
@@ -157,7 +157,7 @@ int wilc_scan(struct wilc_vif *vif, u8 s
 	u32 index = 0;
 	u32 i, scan_timeout;
 	u8 *buffer;
-	u8 valuesize = 0;
+	u32 valuesize = 0;
 	u8 *search_ssid_vals = NULL;
 	struct host_if_drv *hif_drv = vif->hif_drv;
 
author maintainer
#390 lore raw

PATCH 366/491 USB: serial: option: add MeiG Smart SRM825WN

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ernestas Kulik <[email protected]>

commit e8d0ed37bd51da52da6225d278e330c2f18a6198 upstream.

Add support for the SDX62-based MeiG Smart SRM825WN module.

If#= 0: RNDIS
If#= 1: RNDIS
If#= 2: Diag
If#= 3: AT
If#= 4: AT
If#= 5: NMEA

T:  Bus=01 Lev=02 Prnt=02 Port=00 Cnt=01 Dev#= 19 Spd=480  MxCh= 0
D:  Ver= 2.10 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
P:  Vendor=2dee ProdID=4d38 Rev= 5.04
S:  Manufacturer=MEIG
S:  Product=LTE-A Module
S:  SerialNumber=da47a175

C:* #Ifs= 6 Cfg#= 1 Atr=80 MxPwr=500mA
A: FirstIf#= 0 IfCount= 2 Cls=e0(wlcon) Sub=01 Prot=03
I:* If#= 0 Alt= 0 #EPs= 1 Cls=e0(wlcon) Sub=01 Prot=03 Driver=rndis_host
E: Ad=81(I) Atr=03(Int.) MxPS= 8 Ivl=32ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=0a(data ) Sub=00 Prot=00 Driver=rndis_host

E:  Ad=8e(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=0f(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
I:* If#= 2 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=30 Driver=option
E:  Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
I:* If#= 3 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=40 Driver=option
E:  Ad=84(I) Atr=03(Int.) MxPS=  10 Ivl=32ms
E:  Ad=83(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
I:* If#= 4 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=40 Driver=option
E:  Ad=86(I) Atr=03(Int.) MxPS=  10 Ivl=32ms
E:  Ad=85(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=03(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
I:* If#= 5 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=60 Driver=option
E:  Ad=88(I) Atr=03(Int.) MxPS=  10 Ivl=32ms
E:  Ad=87(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=04(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
---
 drivers/usb/serial/option.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/usb/serial/option.c +3 -0
@@ -2441,6 +2441,9 @@ static const struct usb_device_id option
 	{ USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d22, 0xff, 0xff, 0x30) },	/* MeiG Smart SRM815 and SRM825L */
 	{ USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d22, 0xff, 0xff, 0x40) },	/* MeiG Smart SRM825L */
 	{ USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d22, 0xff, 0xff, 0x60) },	/* MeiG Smart SRM825L */
+	{ USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d38, 0xff, 0xff, 0x30) },	/* MeiG Smart SRM825WN (Diag) */
+	{ USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d38, 0xff, 0xff, 0x40) },	/* MeiG Smart SRM825WN (AT) */
+	{ USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d38, 0xff, 0xff, 0x60) },	/* MeiG Smart SRM825WN (NMEA) */
 	{ USB_DEVICE_INTERFACE_CLASS(0x2df3, 0x9d03, 0xff) },			/* LongSung M5710 */
 	{ USB_DEVICE_INTERFACE_CLASS(0x305a, 0x1404, 0xff) },			/* GosunCn GM500 RNDIS */
 	{ USB_DEVICE_INTERFACE_CLASS(0x305a, 0x1405, 0xff) },			/* GosunCn GM500 MBIM */
author maintainer
#391 lore raw

PATCH 367/491 ALSA: caiaq: fix stack out-of-bounds read in init_card

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Berk Cem Goksel <[email protected]>

commit 45424e871abf2a152e247a9cff78359f18dd95c0 upstream.

The loop creates a whitespace-stripped copy of the card shortname where len < sizeof(card->id) is used for the bounds check. Since sizeof(card->id) is 16 and the local id buffer is also 16 bytes, writing 16 non-space characters fills the entire buffer, overwriting the terminating nullbyte.

When this non-null-terminated string is later passed to snd_card_set_id() -> copy_valid_id_string(), the function scans forward with while (*nid && ...) and reads past the end of the stack buffer, reading the contents of the stack.

A USB device with a product name containing many non-ASCII, non-space characters (e.g. multibyte UTF-8) will reliably trigger this as follows:

  BUG: KASAN: stack-out-of-bounds in copy_valid_id_string
       sound/core/init.c:696 [inline]
  BUG: KASAN: stack-out-of-bounds in snd_card_set_id_no_lock+0x698/0x74c
       sound/core/init.c:718

The off-by-one has been present since commit bafeee5b1f8d ("ALSA:
snd_usb_caiaq: give better shortname") from June 2009 (v2.6.31-rc1), which first introduced this whitespace-stripping loop. The original code never accounted for the null terminator when bounding the copy.

Fix this by changing the loop bound to sizeof(card->id) - 1, ensuring at least one byte remains as the null terminator.

---
 sound/usb/caiaq/device.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/usb/caiaq/device.c +1 -1
@@ -502,7 +502,7 @@ static int init_card(struct snd_usb_caia
 		memset(id, 0, sizeof(id));
 
 		for (c = card->shortname, len = 0;
-			*c && len < sizeof(card->id); c++)
+			*c && len < sizeof(card->id) - 1; c++)
 			if (*c != ' ')
 				id[len++] = *c;
 
author maintainer
#392 lore raw

PATCH 368/491 ALSA: ctxfi: Fix missing SPDIFI1 index handling

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Takashi Iwai <[email protected]>

commit b045ab3dff97edae6d538eeff900a34c098761f8 upstream.

SPDIF1 DAIO type isn't properly handled in daio_device_index() for hw20k2, and it returned -EINVAL, which ended up with the out-of-bounds array access. Follow the hw20k1 pattern and return the proper index for this type, too.

---
 sound/pci/ctxfi/ctdaio.c |    1 +
 1 file changed, 1 insertion(+)
sound/pci/ctxfi/ctdaio.c +1 -0
@@ -119,6 +119,7 @@ static unsigned int daio_device_index(en
 		switch (type) {
 		case SPDIFOO:	return 0;
 		case SPDIFIO:	return 0;
+		case SPDIFI1:	return 1;
 		case LINEO1:	return 4;
 		case LINEO2:	return 7;
 		case LINEO3:	return 5;
author maintainer
#393 lore raw

PATCH 369/491 Bluetooth: SMP: derive legacy responder STK authentication from MITM state

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Oleh Konko <[email protected]>

commit 20756fec2f0108cb88e815941f1ffff88dc286fe upstream.

The legacy responder path in smp_random() currently labels the stored STK as authenticated whenever pending_sec_level is BT_SECURITY_HIGH. That reflects what the local service requested, not what the pairing flow actually achieved.

For Just Works/Confirm legacy pairing, SMP_FLAG_MITM_AUTH stays clear and the resulting STK should remain unauthenticated even if the local side requested HIGH security. Use the established MITM state when storing the responder STK so the key metadata matches the pairing result.

This also keeps the legacy path aligned with the Secure Connections code, which already treats JUST_WORKS/JUST_CFM as unauthenticated.

Fixes fff3490f4781 ("Bluetooth: Fix setting correct authentication information for SMP STK")
---
 net/bluetooth/smp.c |    5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
net/bluetooth/smp.c +1 -4
@@ -1018,10 +1018,7 @@ static u8 smp_random(struct smp_chan *sm
 
 		smp_s1(smp->tk, smp->prnd, smp->rrnd, stk);
 
-		if (hcon->pending_sec_level == BT_SECURITY_HIGH)
-			auth = 1;
-		else
-			auth = 0;
+		auth = test_bit(SMP_FLAG_MITM_AUTH, &smp->flags) ? 1 : 0;
 
 		/* Even though there's no _RESPONDER suffix this is the
 		 * responder STK we're adding for later lookup (the initiator
author maintainer
#394 lore raw

PATCH 370/491 Bluetooth: SMP: force responder MITM requirements before building the pairing response

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Oleh Konko <[email protected]>

commit d05111bfe37bfd8bd4d2dfe6675d6bdeef43f7c7 upstream.

smp_cmd_pairing_req() currently builds the pairing response from the initiator auth_req before enforcing the local BT_SECURITY_HIGH requirement. If the initiator omits SMP_AUTH_MITM, the response can also omit it even though the local side still requires MITM.

tk_request() then sees an auth value without SMP_AUTH_MITM and may select JUST_CFM, making method selection inconsistent with the pairing policy the responder already enforces.

When the local side requires HIGH security, first verify that MITM can be achieved from the IO capabilities and then force SMP_AUTH_MITM in the response in both rsp.auth_req and auth. This keeps the responder auth bits and later method selection aligned.

Fixes 2b64d153a0cc ("Bluetooth: Add MITM mechanism to LE-SMP")
---
 net/bluetooth/smp.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
net/bluetooth/smp.c +5 -1
@@ -1818,7 +1818,7 @@ static u8 smp_cmd_pairing_req(struct l2c
 	if (sec_level > conn->hcon->pending_sec_level)
 		conn->hcon->pending_sec_level = sec_level;
 
-	/* If we need MITM check that it can be achieved */
+	/* If we need MITM check that it can be achieved. */
 	if (conn->hcon->pending_sec_level >= BT_SECURITY_HIGH) {
 		u8 method;
 
@@ -1826,6 +1826,10 @@ static u8 smp_cmd_pairing_req(struct l2c
 					 req->io_capability);
 		if (method == JUST_WORKS || method == JUST_CFM)
 			return SMP_AUTH_REQUIREMENTS;
+
+		/* Force MITM bit if it isn't set by the initiator. */
+		auth |= SMP_AUTH_MITM;
+		rsp.auth_req |= SMP_AUTH_MITM;
 	}
 
 	key_size = min(req->max_key_size, rsp.max_key_size);
author maintainer
#395 lore raw

PATCH 371/491 MIPS: Fix the GCC version check for `__multi3 workaround

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maciej W. Rozycki <[email protected]>

commit ec8bf18814915460d9c617b556bf024efef26613 upstream.

It was only GCC 10 that fixed a MIPS64r6 code generation issue with a `__multi3' libcall inefficiently produced to perform 64-bit widening multiplication while suitable machine instructions exist to do such a calculation. The fix went in with GCC commit 48b2123f6336 ("re PR target/82981 (unnecessary __multi3 call for mips64r6 linux kernel)").

Adjust our code accordingly, removing build failures such as:

mips64-linux-ld: lib/math/div64.o: in function `mul_u64_add_u64_div_u64':
div64.c:(.text+0x84): undefined reference to `__multi3'

with the GCC versions affected.

---
 arch/mips/lib/multi3.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
arch/mips/lib/multi3.c +3 -3
@@ -4,12 +4,12 @@
 #include "libgcc.h"
 
 /*
- * GCC 7 & older can suboptimally generate __multi3 calls for mips64r6, so for
+ * GCC 9 & older can suboptimally generate __multi3 calls for mips64r6, so for
  * that specific case only we implement that intrinsic here.
  *
  * See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82981
  */
-#if defined(CONFIG_64BIT) && defined(CONFIG_CPU_MIPSR6) && (__GNUC__ < 8)
+#if defined(CONFIG_64BIT) && defined(CONFIG_CPU_MIPSR6) && (__GNUC__ < 10)
 
 /* multiply 64-bit values, low 64-bits returned */
 static inline long long notrace dmulu(long long a, long long b)
@@ -51,4 +51,4 @@ ti_type notrace __multi3(ti_type a, ti_t
 }
 EXPORT_SYMBOL(__multi3);
 
-#endif /* 64BIT && CPU_MIPSR6 && GCC7 */
+#endif /* 64BIT && CPU_MIPSR6 && GCC9 */
author maintainer
#396 lore raw

PATCH 372/491 hwmon: (occ) Fix division by zero in occ_show_power_1()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sanman Pradhan <[email protected]>

commit 39e2a5bf970402a8530a319cf06122e216ba57b8 upstream.

In occ_show_power_1() case 1, the accumulator is divided by update_tag without checking for zero. If no samples have been collected yet (e.g. during early boot when the sensor block is included but hasn't been updated), update_tag is zero, causing a kernel divide-by-zero crash.

The 2019 fix in commit 211186cae14d ("hwmon: (occ) Fix division by zero issue") only addressed occ_get_powr_avg() used by occ_show_power_2() and occ_show_power_a0(). This separate code path in occ_show_power_1() was missed.

Fix this by reusing the existing occ_get_powr_avg() helper, which already handles the zero-sample case and uses mul_u64_u32_div() to multiply before dividing for better precision. Move the helper above occ_show_power_1() so it is visible at the call site.

Fixes c10e753d43eb ("hwmon (occ): Add sensor types and versions")
[groeck: Fix alignment problems reported by checkpatch]
---
 drivers/hwmon/occ/common.c |   17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)
drivers/hwmon/occ/common.c +8 -9
@@ -422,6 +422,12 @@ static ssize_t occ_show_freq_2(struct de
 	return sysfs_emit(buf, "%u\n", val);
 }
 
+static u64 occ_get_powr_avg(u64 accum, u32 samples)
+{
+	return (samples == 0) ? 0 :
+		mul_u64_u32_div(accum, 1000000UL, samples);
+}
+
 static ssize_t occ_show_power_1(struct device *dev,
 				struct device_attribute *attr, char *buf)
 {
@@ -443,9 +449,8 @@ static ssize_t occ_show_power_1(struct d
 		val = get_unaligned_be16(&power->sensor_id);
 		break;
 	case 1:
-		val = get_unaligned_be32(&power->accumulator) /
-			get_unaligned_be32(&power->update_tag);
-		val *= 1000000ULL;
+		val = occ_get_powr_avg(get_unaligned_be32(&power->accumulator),
+				       get_unaligned_be32(&power->update_tag));
 		break;
 	case 2:
 		val = (u64)get_unaligned_be32(&power->update_tag) *
@@ -461,12 +466,6 @@ static ssize_t occ_show_power_1(struct d
 	return sysfs_emit(buf, "%llu\n", val);
 }
 
-static u64 occ_get_powr_avg(u64 accum, u32 samples)
-{
-	return (samples == 0) ? 0 :
-		mul_u64_u32_div(accum, 1000000UL, samples);
-}
-
 static ssize_t occ_show_power_2(struct device *dev,
 				struct device_attribute *attr, char *buf)
 {
author maintainer
#397 lore raw

PATCH 373/491 drm/ast: dp501: Fix initialization of SCU2C

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thomas Zimmermann <[email protected]>

commit 2f42c1a6161646cbd29b443459fd635d29eda634 upstream.

Ast's DP501 initialization reads the register SCU2C at offset 0x1202c and tries to set it to source data from VGA. But writes the update to offset 0x0, with unknown results. Write the result to SCU instead.

The bug only happens in ast_init_analog(). There's similar code in ast_init_dvo(), which works correctly.

Fixes 83c6620bae3f ("drm/ast: initial DP501 support (v0.2)")
Cc Dave Airlie <[email protected]>
Cc Thomas Zimmermann <[email protected]>
Cc Jocelyn Falempe <[email protected]>
Cc <[email protected]> # v3.16+
---
 drivers/gpu/drm/ast/ast_dp501.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/ast/ast_dp501.c +1 -1
@@ -484,7 +484,7 @@ static void ast_init_analog(struct drm_d
 	/* Finally, clear bits [17:16] of SCU2c */
 	data = ast_read32(ast, 0x1202c);
 	data &= 0xfffcffff;
-	ast_write32(ast, 0, data);
+	ast_write32(ast, 0x1202c, data);
 
 	/* Disable DVO */
 	ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xa3, 0xcf, 0x00);
author maintainer
#398 lore raw

PATCH 374/491 USB: serial: io_edgeport: add support for Blackbox IC135A

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Frej Drejhammar <[email protected]>

commit 0e01c3416eb863ee7f156a9d7e7421ec0a9f68a0 upstream.

The Blackbox 724-746-5500 USB Director USB-RS-232 HUB, part number IC135A, is a rebadged Edgeport/4 with its own USB device id.

---
 drivers/usb/serial/io_edgeport.c |    3 +++
 drivers/usb/serial/io_usbvend.h  |    1 +
 2 files changed, 4 insertions(+)
drivers/usb/serial/io_edgeport.c +3 -0
@@ -73,6 +73,7 @@ static const struct usb_device_id edgepo
 	{ USB_DEVICE(USB_VENDOR_ID_ION, ION_DEVICE_ID_EDGEPORT_22I) },
 	{ USB_DEVICE(USB_VENDOR_ID_ION, ION_DEVICE_ID_EDGEPORT_412_4) },
 	{ USB_DEVICE(USB_VENDOR_ID_ION, ION_DEVICE_ID_EDGEPORT_COMPATIBLE) },
+	{ USB_DEVICE(USB_VENDOR_ID_ION, ION_DEVICE_ID_BLACKBOX_IC135A) },
 	{ }
 };
 
@@ -121,6 +122,7 @@ static const struct usb_device_id id_tab
 	{ USB_DEVICE(USB_VENDOR_ID_ION, ION_DEVICE_ID_EDGEPORT_8R) },
 	{ USB_DEVICE(USB_VENDOR_ID_ION, ION_DEVICE_ID_EDGEPORT_8RR) },
 	{ USB_DEVICE(USB_VENDOR_ID_ION, ION_DEVICE_ID_EDGEPORT_412_8) },
+	{ USB_DEVICE(USB_VENDOR_ID_ION, ION_DEVICE_ID_BLACKBOX_IC135A) },
 	{ USB_DEVICE(USB_VENDOR_ID_NCR, NCR_DEVICE_ID_EPIC_0202) },
 	{ USB_DEVICE(USB_VENDOR_ID_NCR, NCR_DEVICE_ID_EPIC_0203) },
 	{ USB_DEVICE(USB_VENDOR_ID_NCR, NCR_DEVICE_ID_EPIC_0310) },
@@ -546,6 +548,7 @@ static void get_product_info(struct edge
 	case ION_DEVICE_ID_EDGEPORT_2_DIN:
 	case ION_DEVICE_ID_EDGEPORT_4_DIN:
 	case ION_DEVICE_ID_EDGEPORT_16_DUAL_CPU:
+	case ION_DEVICE_ID_BLACKBOX_IC135A:
 		product_info->IsRS232 = 1;
 		break;
 
drivers/usb/serial/io_usbvend.h +1 -0
@@ -211,6 +211,7 @@
 
 //
 // Definitions for other product IDs
+#define ION_DEVICE_ID_BLACKBOX_IC135A		0x0801	// OEM device (rebranded Edgeport/4)
 #define ION_DEVICE_ID_MT4X56USB			0x1403	// OEM device
 #define ION_DEVICE_ID_E5805A			0x1A01  // OEM device (rebranded Edgeport/4)
 
author maintainer
#399 lore raw

PATCH 375/491 USB: serial: option: add support for Rolling Wireless RW135R-GL

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wanquan Zhong <[email protected]>

commit 01e8d0f742222f1e68f48180d5480097adf7ae9f upstream.

Add VID/PID 33f8:1003 for the Rolling Wireless RW135R-GL M.2 module, which is used in laptop debug cards with MBIM interface for Linux/Chrome OS. The device supports mbim, pipe functionalities.

Here are the outputs of usb-devices:

T:  Bus=04 Lev=01 Prnt=01 Port=02 Cnt=01 Dev#=  2 Spd=5000 MxCh= 0
D:  Ver= 3.20 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 9 #Cfgs=  1
P:  Vendor=33f8 ProdID=1003 Rev=05.15
S:  Manufacturer=Rolling Wireless S.a.r.l.
S:  Product=Rolling RW135R-GL Module
S:  SerialNumber=12345678
C:  #Ifs= 3 Cfg#= 1 Atr=a0 MxPwr=896mA
I:  If#= 0 Alt= 0 #EPs= 1 Cls=02(commc) Sub=0e Prot=00 Driver=cdc_mbim
E:  Ad=81(I) Atr=03(Int.) MxPS=  64 Ivl=32ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=0a(data ) Sub=00 Prot=02 Driver=cdc_mbim
E:  Ad=0f(O) Atr=02(Bulk) MxPS=1024 Ivl=0ms
E:  Ad=8e(I) Atr=02(Bulk) MxPS=1024 Ivl=0ms
I:  If#= 2 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=00 Prot=40 Driver=option
E:  Ad=01(O) Atr=02(Bulk) MxPS=1024 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=1024 Ivl=0ms
E:  Ad=83(I) Atr=03(Int.) MxPS=  10 Ivl=32ms
---
 drivers/usb/serial/option.c |    1 +
 1 file changed, 1 insertion(+)
drivers/usb/serial/option.c +1 -0
@@ -2464,6 +2464,7 @@ static const struct usb_device_id option
 	{ USB_DEVICE_INTERFACE_CLASS(0x33f8, 0x0302, 0xff) },			/* Rolling RW101R-GL (laptop MBIM) */
 	{ USB_DEVICE_INTERFACE_CLASS(0x33f8, 0x0802, 0xff),			/* Rolling RW350-GL (laptop MBIM) */
 	  .driver_info = RSVD(5) },
+	{ USB_DEVICE_INTERFACE_CLASS(0x33f8, 0x1003, 0xff) },			/* Rolling RW135R-GL (laptop MBIM) */
 	{ USB_DEVICE_AND_INTERFACE_INFO(0x3731, 0x0100, 0xff, 0xff, 0x30) },	/* NetPrisma LCUK54-WWD for Global */
 	{ USB_DEVICE_AND_INTERFACE_INFO(0x3731, 0x0100, 0xff, 0x00, 0x40) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(0x3731, 0x0100, 0xff, 0xff, 0x40) },
author maintainer
#400 lore raw

PATCH 376/491 USB: core: add NO_LPM quirk for Razer Kiyo Pro webcam

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: JP Hein <[email protected]>

commit 8b7a42ecdcdeb55580d9345412f7f8fc5aca3f6c upstream.

The Razer Kiyo Pro (1532:0e05) is a USB 3.0 UVC webcam whose firmware does not handle USB Link Power Management transitions reliably. When LPM is active, the device can enter a state where it fails to respond to control transfers, producing EPIPE (-32) errors on UVC probe control SET_CUR requests. In the worst case, the stalled endpoint triggers an xHCI stop-endpoint command that times out, causing the host controller to be declared dead and every USB device on the bus to be disconnected.

This has been reported as Ubuntu Launchpad Bug #2061177. The failure mode is:

1. UVC probe control SET_CUR returns -32 (EPIPE)
2. xHCI host not responding to stop endpoint command
3. xHCI host controller not responding, assume dead
4. All USB devices on the affected xHCI controller disconnect

Disabling LPM prevents the firmware from entering the problematic low- power states that precede the stall. This is the same approach used for other webcams with similar firmware issues (e.g., Logitech HD Webcam C270).

---
 drivers/usb/core/quirks.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/usb/core/quirks.c +2 -0
@@ -477,6 +477,8 @@ static const struct usb_device_id usb_qu
 	/* Razer - Razer Blade Keyboard */
 	{ USB_DEVICE(0x1532, 0x0116), .driver_info =
 			USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL },
+	/* Razer - Razer Kiyo Pro Webcam */
+	{ USB_DEVICE(0x1532, 0x0e05), .driver_info = USB_QUIRK_NO_LPM },
 
 	/* Lenovo ThinkPad OneLink+ Dock twin hub controllers (VIA Labs VL812) */
 	{ USB_DEVICE(0x17ef, 0x1018), .driver_info = USB_QUIRK_RESET_RESUME },
author maintainer
#401 lore raw

PATCH 377/491 Input: synaptics-rmi4 - fix a locking bug in an error path

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bart Van Assche <[email protected]>

commit 7adaaee5edd35a423ae199c41b86bd1ed60ed483 upstream.

Lock f54->data_mutex when entering the function statement since jumping to the 'error' label when checking report_size fails causes that mutex to be unlocked.

This bug has been detected by the Clang thread-safety checker.

Fixes 3a762dbd5347 ("[media] Input: synaptics-rmi4 - add support for F54 diagnostics")
---
 drivers/input/rmi4/rmi_f54.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/input/rmi4/rmi_f54.c +2 -2
@@ -534,6 +534,8 @@ static void rmi_f54_work(struct work_str
 	int error;
 	int i;
 
+	mutex_lock(&f54->data_mutex);
+
 	report_size = rmi_f54_get_report_size(f54);
 	if (report_size == 0) {
 		dev_err(&fn->dev, "Bad report size, report type=%d\n",
@@ -542,8 +544,6 @@ static void rmi_f54_work(struct work_str
 		goto error;     /* retry won't help */
 	}
 
-	mutex_lock(&f54->data_mutex);
-
 	/*
 	 * Need to check if command has completed.
 	 * If not try again later.
author maintainer
#402 lore raw

PATCH 378/491 Input: i8042 - add TUXEDO InfinityBook Max 16 Gen10 AMD to i8042 quirk table

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christoffer Sandberg <[email protected]>

commit 5839419cffc7788a356428d321e3ec18055c0286 upstream.

The device occasionally wakes up from suspend with missing input on the internal keyboard and the following suspend attempt results in an instant wake-up. The quirks fix both issues for this device.

---
 drivers/input/serio/i8042-acpipnpio.h |    7 +++++++
 1 file changed, 7 insertions(+)
drivers/input/serio/i8042-acpipnpio.h +7 -0
@@ -1189,6 +1189,13 @@ static const struct dmi_system_id i8042_
 	},
 	{
 		.matches = {
+			DMI_MATCH(DMI_BOARD_NAME, "X6KK45xU_X6SP45xU"),
+		},
+		.driver_data = (void *)(SERIO_QUIRK_NOMUX | SERIO_QUIRK_RESET_ALWAYS |
+					SERIO_QUIRK_NOLOOP | SERIO_QUIRK_NOPNP)
+	},
+	{
+		.matches = {
 			DMI_MATCH(DMI_BOARD_NAME, "WUJIE Series-X5SP4NAG"),
 		},
 		.driver_data = (void *)(SERIO_QUIRK_NOMUX | SERIO_QUIRK_RESET_ALWAYS |
author maintainer
#403 lore raw

PATCH 379/491 Input: xpad - add support for Razer Wolverine V3 Pro

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zoltan Illes <[email protected]>

commit e2b0ae529db4766584e77647cefe3ec15c3d842e upstream.

Add device IDs for the Razer Wolverine V3 Pro controller in both wired (0x0a57) and wireless 2.4 GHz dongle (0x0a59) modes.

The controller uses the Xbox 360 protocol (vendor-specific class, subclass 93, protocol 1) on interface 0 with an identical 20-byte input report layout, so no additional processing is needed.

---
 drivers/input/joystick/xpad.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/input/joystick/xpad.c +2 -0
@@ -269,6 +269,8 @@ static const struct xpad_device {
 	{ 0x1532, 0x0a00, "Razer Atrox Arcade Stick", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOXONE },
 	{ 0x1532, 0x0a03, "Razer Wildcat", 0, XTYPE_XBOXONE },
 	{ 0x1532, 0x0a29, "Razer Wolverine V2", 0, XTYPE_XBOXONE },
+	{ 0x1532, 0x0a57, "Razer Wolverine V3 Pro (Wired)", 0, XTYPE_XBOX360 },
+	{ 0x1532, 0x0a59, "Razer Wolverine V3 Pro (2.4 GHz Dongle)", 0, XTYPE_XBOX360 },
 	{ 0x15e4, 0x3f00, "Power A Mini Pro Elite", 0, XTYPE_XBOX360 },
 	{ 0x15e4, 0x3f0a, "Xbox Airflo wired controller", 0, XTYPE_XBOX360 },
 	{ 0x15e4, 0x3f10, "Batarang Xbox 360 controller", 0, XTYPE_XBOX360 },
author maintainer
#404 lore raw

PATCH 380/491 iio: dac: ad5770r: fix error return in ad5770r_read_raw()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Antoniu Miclaus <[email protected]>

commit c354521708175d776d896f8bdae44b18711eccb6 upstream.

Return the error code from regmap_bulk_read() instead of 0 so that I/O failures are properly propagated.

Fixes cbbb819837f6 ("iio: dac: ad5770r: Add AD5770R support")
---
 drivers/iio/dac/ad5770r.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/dac/ad5770r.c +1 -1
@@ -323,7 +323,7 @@ static int ad5770r_read_raw(struct iio_d
 				       chan->address,
 				       st->transf_buf, 2);
 		if (ret)
-			return 0;
+			return ret;
 
 		buf16 = st->transf_buf[0] + (st->transf_buf[1] << 8);
 		*val = buf16 >> 2;
author maintainer
#405 lore raw

PATCH 381/491 iio: light: vcnl4035: fix scan buffer on big-endian

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Lechner <[email protected]>

commit fdc7aa54a5d44c05880a4aad7cfb41aacfd16d7b upstream.

Rework vcnl4035_trigger_consumer_handler() so that we are not passing what should be a u16 value as an int * to regmap_read(). This won't work on bit endian systems.

Instead, add a new unsigned int variable to pass to regmap_read(). Then copy that value into the buffer struct.

The buffer array is replaced with a struct since there is only one value being read. This allows us to use the correct u16 data type and has a side-effect of simplifying the alignment specification.

Also fix the endianness of the scan format from little-endian to CPU endianness. Since we are using regmap to read the value, it will be CPU-endian.

Fixes 55707294c4eb ("iio: light: Add support for vishay vcnl4035")
---
 drivers/iio/light/vcnl4035.c |   18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)
drivers/iio/light/vcnl4035.c +12 -6
@@ -105,17 +105,23 @@ static irqreturn_t vcnl4035_trigger_cons
 	struct iio_dev *indio_dev = pf->indio_dev;
 	struct vcnl4035_data *data = iio_priv(indio_dev);
 	/* Ensure naturally aligned timestamp */
-	u8 buffer[ALIGN(sizeof(u16), sizeof(s64)) + sizeof(s64)]  __aligned(8) = { };
+	struct {
+		u16 als_data;
+		aligned_s64 timestamp;
+	} buffer = { };
+	unsigned int val;
 	int ret;
 
-	ret = regmap_read(data->regmap, VCNL4035_ALS_DATA, (int *)buffer);
+	ret = regmap_read(data->regmap, VCNL4035_ALS_DATA, &val);
 	if (ret < 0) {
 		dev_err(&data->client->dev,
 			"Trigger consumer can't read from sensor.\n");
 		goto fail_read;
 	}
-	iio_push_to_buffers_with_timestamp(indio_dev, buffer,
-					iio_get_time_ns(indio_dev));
+
+	buffer.als_data = val;
+	iio_push_to_buffers_with_timestamp(indio_dev, &buffer,
+					   iio_get_time_ns(indio_dev));
 
 fail_read:
 	iio_trigger_notify_done(indio_dev->trig);
@@ -380,7 +386,7 @@ static const struct iio_chan_spec vcnl40
 			.sign = 'u',
 			.realbits = 16,
 			.storagebits = 16,
-			.endianness = IIO_LE,
+			.endianness = IIO_CPU,
 		},
 	},
 	{
@@ -394,7 +400,7 @@ static const struct iio_chan_spec vcnl40
 			.sign = 'u',
 			.realbits = 16,
 			.storagebits = 16,
-			.endianness = IIO_LE,
+			.endianness = IIO_CPU,
 		},
 	},
 };
author maintainer
#406 lore raw

PATCH 382/491 iio: imu: st_lsm6dsx: Set FIFO ODR for accelerometer and gyroscope only

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Francesco Lavra <[email protected]>

commit 630748afa7030b272b7bee5df857e7bcf132ed51 upstream.

The st_lsm6dsx_set_fifo_odr() function, which is called when enabling and disabling the hardware FIFO, checks the contents of the hw->settings->batch array at index sensor->id, and then sets the current ODR value in sensor registers that depend on whether the register address is set in the above array element. This logic is valid for internal sensors only, i.e. the accelerometer and gyroscope; however, since commit c91c1c844ebd ("iio: imu:
st_lsm6dsx: add i2c embedded controller support"), this function is called also when configuring the hardware FIFO for external sensors (i.e. sensors accessed through the sensor hub functionality), which can result in unrelated device registers being written.

Add a check to the beginning of st_lsm6dsx_set_fifo_odr() so that it does not touch any registers unless it is called for internal sensors.

Fixes c91c1c844ebd ("iio: imu: st_lsm6dsx: add i2c embedded controller support")
---
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c |    4 ++++
 1 file changed, 4 insertions(+)
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c +4 -0
@@ -201,6 +201,10 @@ static int st_lsm6dsx_set_fifo_odr(struc
 	const struct st_lsm6dsx_reg *batch_reg;
 	u8 data;
 
+	/* Only internal sensors have a FIFO ODR configuration register. */
+	if (sensor->id >= ARRAY_SIZE(hw->settings->batch))
+		return 0;
+
 	batch_reg = &hw->settings->batch[sensor->id];
 	if (batch_reg->addr) {
 		int val;
author maintainer
#407 lore raw

PATCH 383/491 iio: gyro: mpu3050: Fix incorrect free_irq() variable

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ethan Tidmore <[email protected]>

commit edb11a1aef4011a4b7b22cc3c3396c6fe371f4a6 upstream.

The handler for the IRQ part of this driver is mpu3050->trig but, in the teardown free_irq() is called with handler mpu3050.

Use correct IRQ handler when calling free_irq().

Fixes 3904b28efb2c7 ("iio: gyro: Add driver for the MPU-3050 gyroscope")
---
 drivers/iio/gyro/mpu3050-core.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/gyro/mpu3050-core.c +1 -1
@@ -1278,7 +1278,7 @@ int mpu3050_common_remove(struct device
 	pm_runtime_disable(dev);
 	iio_triggered_buffer_cleanup(indio_dev);
 	if (mpu3050->irq)
-		free_irq(mpu3050->irq, mpu3050);
+		free_irq(mpu3050->irq, mpu3050->trig);
 	iio_device_unregister(indio_dev);
 	mpu3050_power_down(mpu3050);
 
author maintainer
#408 lore raw

PATCH 384/491 iio: gyro: mpu3050: Fix irq resource leak

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ethan Tidmore <[email protected]>

commit 4216db1043a3be72ef9c2b7b9f393d7fa72496e6 upstream.

The interrupt handler is setup but only a few lines down if iio_trigger_register() fails the function returns without properly releasing the handler.

Add cleanup goto to resolve resource leak.

Detected by Smatch:
drivers/iio/gyro/mpu3050-core.c:1128 mpu3050_trigger_probe() warn:
'irq' from request_threaded_irq() not released on lines: 1124.

Fixes 3904b28efb2c7 ("iio: gyro: Add driver for the MPU-3050 gyroscope")
---
 drivers/iio/gyro/mpu3050-core.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/iio/gyro/mpu3050-core.c +6 -1
@@ -1137,11 +1137,16 @@ static int mpu3050_trigger_probe(struct
 
 	ret = iio_trigger_register(mpu3050->trig);
 	if (ret)
-		return ret;
+		goto err_iio_trigger;
 
 	indio_dev->trig = iio_trigger_get(mpu3050->trig);
 
 	return 0;
+
+err_iio_trigger:
+	free_irq(mpu3050->irq, mpu3050->trig);
+
+	return ret;
 }
 
 int mpu3050_common_probe(struct device *dev,
author maintainer
#409 lore raw

PATCH 385/491 iio: gyro: mpu3050: Move iio_device_register() to correct location

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ethan Tidmore <[email protected]>

commit 4c05799449108fb0e0a6bd30e65fffc71e60db4d upstream.

iio_device_register() should be at the end of the probe function to prevent race conditions.

Place iio_device_register() at the end of the probe function and place iio_device_unregister() accordingly.

---
 drivers/iio/gyro/mpu3050-core.c |   21 +++++++++++++--------
 1 file changed, 13 insertions(+), 8 deletions(-)
drivers/iio/gyro/mpu3050-core.c +13 -8
@@ -1234,12 +1234,6 @@ int mpu3050_common_probe(struct device *
 		goto err_power_down;
 	}
 
-	ret = iio_device_register(indio_dev);
-	if (ret) {
-		dev_err(dev, "device register failed\n");
-		goto err_cleanup_buffer;
-	}
-
 	dev_set_drvdata(dev, indio_dev);
 
 	/* Check if we have an assigned IRQ to use as trigger */
@@ -1262,9 +1256,20 @@ int mpu3050_common_probe(struct device *
 	pm_runtime_use_autosuspend(dev);
 	pm_runtime_put(dev);
 
+	ret = iio_device_register(indio_dev);
+	if (ret) {
+		dev_err(dev, "device register failed\n");
+		goto err_iio_device_register;
+	}
+
 	return 0;
 
-err_cleanup_buffer:
+err_iio_device_register:
+	pm_runtime_get_sync(dev);
+	pm_runtime_put_noidle(dev);
+	pm_runtime_disable(dev);
+	if (irq)
+		free_irq(mpu3050->irq, mpu3050->trig);
 	iio_triggered_buffer_cleanup(indio_dev);
 err_power_down:
 	mpu3050_power_down(mpu3050);
@@ -1278,13 +1283,13 @@ int mpu3050_common_remove(struct device
 	struct iio_dev *indio_dev = dev_get_drvdata(dev);
 	struct mpu3050 *mpu3050 = iio_priv(indio_dev);
 
+	iio_device_unregister(indio_dev);
 	pm_runtime_get_sync(dev);
 	pm_runtime_put_noidle(dev);
 	pm_runtime_disable(dev);
 	iio_triggered_buffer_cleanup(indio_dev);
 	if (mpu3050->irq)
 		free_irq(mpu3050->irq, mpu3050->trig);
-	iio_device_unregister(indio_dev);
 	mpu3050_power_down(mpu3050);
 
 	return 0;
author maintainer
#410 lore raw

PATCH 386/491 iio: gyro: mpu3050: Fix out-of-sequence free_irq()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ethan Tidmore <[email protected]>

commit d14116f6529fa085b1a1b1f224dc9604e4d2a29c upstream.

The triggered buffer is initialized before the IRQ is requested. The removal path currently calls iio_triggered_buffer_cleanup() before free_irq(). This violates the expected LIFO.

Place free_irq() in the correct location relative to iio_triggered_buffer_cleanup().

---
 drivers/iio/gyro/mpu3050-core.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/gyro/mpu3050-core.c +1 -1
@@ -1287,9 +1287,9 @@ int mpu3050_common_remove(struct device
 	pm_runtime_get_sync(dev);
 	pm_runtime_put_noidle(dev);
 	pm_runtime_disable(dev);
-	iio_triggered_buffer_cleanup(indio_dev);
 	if (mpu3050->irq)
 		free_irq(mpu3050->irq, mpu3050->trig);
+	iio_triggered_buffer_cleanup(indio_dev);
 	mpu3050_power_down(mpu3050);
 
 	return 0;
author maintainer
#411 lore raw

PATCH 387/491 usb: quirks: add DELAY_INIT quirk for another Silicon Motion flash drive

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miao Li <[email protected]>

commit dd36014ec6042f424ef51b923e607772f7502ee7 upstream.

Another Silicon Motion flash drive also randomly work incorrectly (lsusb does not list the device) on Huawei hisi platforms during 500 reboot cycles, and the DELAY_INIT quirk fixes this issue.

---
 drivers/usb/core/quirks.c |    1 +
 1 file changed, 1 insertion(+)
drivers/usb/core/quirks.c +1 -0
@@ -394,6 +394,7 @@ static const struct usb_device_id usb_qu
 
 	/* Silicon Motion Flash Drive */
 	{ USB_DEVICE(0x090c, 0x1000), .driver_info = USB_QUIRK_DELAY_INIT },
+	{ USB_DEVICE(0x090c, 0x2000), .driver_info = USB_QUIRK_DELAY_INIT },
 
 	/* Sound Devices USBPre2 */
 	{ USB_DEVICE(0x0926, 0x0202), .driver_info =
author maintainer
#412 lore raw

PATCH 388/491 usb: ulpi: fix double free in ulpi_register_interface() error path

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Guangshuo Li <[email protected]>

commit 01af542392b5d41fd659d487015a71f627accce3 upstream.

When device_register() fails, ulpi_register() calls put_device() on ulpi->dev.

The device release callback ulpi_dev_release() drops the OF node reference and frees ulpi, but the current error path in ulpi_register_interface() then calls kfree(ulpi) again, causing a double free.

Let put_device() handle the cleanup through ulpi_dev_release() and avoid freeing ulpi again in ulpi_register_interface().

---
 drivers/usb/common/ulpi.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/usb/common/ulpi.c +2 -3
@@ -288,10 +288,9 @@ struct ulpi *ulpi_register_interface(str
 	ulpi->ops = ops;
 
 	ret = ulpi_register(dev, ulpi);
-	if (ret) {
-		kfree(ulpi);
+	if (ret)
 		return ERR_PTR(ret);
-	}
+
 
 	return ulpi;
 }
author maintainer
#413 lore raw

PATCH 389/491 usb: usbtmc: Flush anchored URBs in usbtmc_release

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Heitor Alves de Siqueira <[email protected]>

commit 8a768552f7a8276fb9e01d49773d2094ace7c8f1 upstream.

When calling usbtmc_release, pending anchored URBs must be flushed or killed to prevent use-after-free errors (e.g. in the HCD giveback path). Call usbtmc_draw_down() to allow anchored URBs to be completed.

---
 drivers/usb/class/usbtmc.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/usb/class/usbtmc.c +3 -0
@@ -254,6 +254,9 @@ static int usbtmc_release(struct inode *
 	list_del(&file_data->file_elem);
 
 	spin_unlock_irq(&file_data->data->dev_lock);
+
+	/* flush anchored URBs */
+	usbtmc_draw_down(file_data);
 	mutex_unlock(&file_data->data->io_mutex);
 
 	kref_put(&file_data->data->kref, usbtmc_delete);
author maintainer
#414 lore raw

PATCH 390/491 usb: ehci-brcm: fix sleep during atomic

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Justin Chen <[email protected]>

commit 679b771ea05ad0f8eeae83e14a91b8f4f39510c4 upstream.

echi_brcm_wait_for_sof() gets called after disabling interrupts in ehci_brcm_hub_control(). Use the atomic version of poll_timeout to fix the warning.

Fixes 9df231511bd6 ("usb: ehci: Add new EHCI driver for Broadcom STB SoC's")
Cc stable <[email protected]>
---
 drivers/usb/host/ehci-brcm.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/usb/host/ehci-brcm.c +2 -2
@@ -31,8 +31,8 @@ static inline void ehci_brcm_wait_for_so
 	int res;
 
 	/* Wait for next microframe (every 125 usecs) */
-	res = readl_relaxed_poll_timeout(&ehci->regs->frame_index, val,
-					 val != frame_idx, 1, 130);
+	res = readl_relaxed_poll_timeout_atomic(&ehci->regs->frame_index,
+						val, val != frame_idx, 1, 130);
 	if (res)
 		ehci_err(ehci, "Error waiting for SOF\n");
 	udelay(delay);
author maintainer
#415 lore raw

PATCH 391/491 phy: renesas: rcar-gen3-usb2: Fix role detection on unbind/bind

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Claudiu Beznea <[email protected]>

commit 54c4c58713aaff76c2422ff5750e557ab3b100d7 upstream.

It has been observed on the Renesas RZ/G3S SoC that unbinding and binding the PHY driver leads to role autodetection failures. This issue occurs when PHY 3 is the first initialized PHY. PHY 3 does not have an interrupt associated with the USB2_INT_ENABLE register (as rcar_gen3_int_enable[3] = 0). As a result, rcar_gen3_init_otg() is called to initialize OTG without enabling PHY interrupts.

To resolve this, add rcar_gen3_is_any_otg_rphy_initialized() and call it in role_store(), role_show(), and rcar_gen3_init_otg(). At the same time, rcar_gen3_init_otg() is only called when initialization for a PHY with interrupt bits is in progress. As a result, the
struct rcar_gen3_phy::otg_initialized is no longer needed.

[claudiu.beznea: declare the i iterrator from
rcar_gen3_is_any_otg_rphy_initialized() outside of for loop]

---
 drivers/phy/renesas/phy-rcar-gen3-usb2.c | 32 +++++++++++-------------
 1 file changed, 14 insertions(+), 18 deletions(-)
drivers/phy/renesas/phy-rcar-gen3-usb2.c +14 -18
index ea01a121b8fc5..646a5140b30e1 100644
@@ -98,7 +98,6 @@ struct rcar_gen3_phy {
 	struct rcar_gen3_chan *ch;
 	u32 int_enable_bits;
 	bool initialized;
-	bool otg_initialized;
 	bool powered;
 };
 
@@ -288,16 +287,16 @@ static bool rcar_gen3_is_any_rphy_initialized(struct rcar_gen3_chan *ch)
 	return false;
 }
 
-static bool rcar_gen3_needs_init_otg(struct rcar_gen3_chan *ch)
+static bool rcar_gen3_is_any_otg_rphy_initialized(struct rcar_gen3_chan *ch)
 {
-	int i;
+	enum rcar_gen3_phy_index i;
 
-	for (i = 0; i < NUM_OF_PHYS; i++) {
-		if (ch->rphys[i].otg_initialized)
-			return false;
+	for (i = PHY_INDEX_BOTH_HC; i <= PHY_INDEX_EHCI; i++) {
+		if (ch->rphys[i].initialized)
+			return true;
 	}
 
-	return true;
+	return false;
 }
 
 static bool rcar_gen3_are_all_rphys_power_off(struct rcar_gen3_chan *ch)
@@ -319,7 +318,7 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
 	bool is_b_device;
 	enum phy_mode cur_mode, new_mode;
 
-	if (!ch->is_otg_channel || !rcar_gen3_is_any_rphy_initialized(ch))
+	if (!ch->is_otg_channel || !rcar_gen3_is_any_otg_rphy_initialized(ch))
 		return -EIO;
 
 	if (sysfs_streq(buf, "host"))
@@ -357,7 +356,7 @@ static ssize_t role_show(struct device *dev, struct device_attribute *attr,
 {
 	struct rcar_gen3_chan *ch = dev_get_drvdata(dev);
 
-	if (!ch->is_otg_channel || !rcar_gen3_is_any_rphy_initialized(ch))
+	if (!ch->is_otg_channel || !rcar_gen3_is_any_otg_rphy_initialized(ch))
 		return -EIO;
 
 	return sprintf(buf, "%s\n", rcar_gen3_is_host(ch) ? "host" :
@@ -370,6 +369,9 @@ static void rcar_gen3_init_otg(struct rcar_gen3_chan *ch)
 	void __iomem *usb2_base = ch->base;
 	u32 val;
 
+	if (!ch->is_otg_channel || rcar_gen3_is_any_otg_rphy_initialized(ch))
+		return;
+
 	/* Should not use functions of read-modify-write a register */
 	val = readl(usb2_base + USB2_LINECTRL1);
 	val = (val & ~USB2_LINECTRL1_DP_RPD) | USB2_LINECTRL1_DPRPD_EN |
@@ -435,12 +437,9 @@ static int rcar_gen3_phy_usb2_init(struct phy *p)
 		writel(USB2_OC_TIMSET_INIT, usb2_base + USB2_OC_TIMSET);
 	}
 
-	/* Initialize otg part */
-	if (channel->is_otg_channel) {
-		if (rcar_gen3_needs_init_otg(channel))
-			rcar_gen3_init_otg(channel);
-		rphy->otg_initialized = true;
-	}
+	/* Initialize otg part (only if we initialize a PHY with IRQs). */
+	if (rphy->int_enable_bits)
+		rcar_gen3_init_otg(channel);
 
 	rphy->initialized = true;
 
@@ -456,9 +455,6 @@ static int rcar_gen3_phy_usb2_exit(struct phy *p)
 
 	rphy->initialized = false;
 
-	if (channel->is_otg_channel)
-		rphy->otg_initialized = false;
-
 	val = readl(usb2_base + USB2_INT_ENABLE);
 	val &= ~rphy->int_enable_bits;
 	if (!rcar_gen3_is_any_rphy_initialized(channel))
Signature · 1 line
2.53.0
author maintainer
#416 lore raw

PATCH 392/491 phy: renesas: rcar-gen3-usb2: Move IRQ request in probe

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Claudiu Beznea <[email protected]>

commit de76809f60cc938d3580bbbd5b04b7d12af6ce3a upstream.

Commit 08b0ad375ca6 ("phy: renesas: rcar-gen3-usb2: move IRQ registration to init") moved the IRQ request operation from probe to
struct phy_ops::phy_init API to avoid triggering interrupts (which lead to register accesses) while the PHY clocks (enabled through runtime PM APIs) are not active. If this happens, it results in a synchronous abort.

One way to reproduce this issue is by enabling CONFIG_DEBUG_SHIRQ, which calls free_irq() on driver removal.

Move the IRQ request and free operations back to probe, and take the runtime PM state into account in IRQ handler. This commit is preparatory for the subsequent fixes in this series.

[claudiu.beznea: fixed conflicts by:
- dropping irq and obint_enable_bits members of rcar_gen3_chan
- using USB2_OBINT_BITS marco in rcar_gen3_phy_usb2_irq()
- keeping irq local variable in rcar_gen3_phy_usb2_probe()
- dropping channel->irq and channel->obint_enable_bits asssignement from probe
- keeping platform_set_drvdata() and channel->dev assignment in probe]

---
 drivers/phy/renesas/phy-rcar-gen3-usb2.c | 46 +++++++++++++-----------
 1 file changed, 26 insertions(+), 20 deletions(-)
drivers/phy/renesas/phy-rcar-gen3-usb2.c +26 -20
index 646a5140b30e1..f66e0daa23648 100644
@@ -110,7 +110,6 @@ struct rcar_gen3_chan {
 	struct work_struct work;
 	struct mutex lock;	/* protects rphys[...].powered */
 	enum usb_dr_mode dr_mode;
-	int irq;
 	bool extcon_host;
 	bool is_otg_channel;
 	bool uses_otg_pins;
@@ -396,16 +395,25 @@ static irqreturn_t rcar_gen3_phy_usb2_irq(int irq, void *_ch)
 {
 	struct rcar_gen3_chan *ch = _ch;
 	void __iomem *usb2_base = ch->base;
-	u32 status = readl(usb2_base + USB2_OBINTSTA);
+	struct device *dev = ch->dev;
 	irqreturn_t ret = IRQ_NONE;
+	u32 status;
 
+	pm_runtime_get_noresume(dev);
+
+	if (pm_runtime_suspended(dev))
+		goto rpm_put;
+
+	status = readl(usb2_base + USB2_OBINTSTA);
 	if (status & USB2_OBINT_BITS) {
-		dev_vdbg(ch->dev, "%s: %08x\n", __func__, status);
+		dev_vdbg(dev, "%s: %08x\n", __func__, status);
 		writel(USB2_OBINT_BITS, usb2_base + USB2_OBINTSTA);
 		rcar_gen3_device_recognition(ch);
 		ret = IRQ_HANDLED;
 	}
 
+rpm_put:
+	pm_runtime_put_noidle(dev);
 	return ret;
 }
 
@@ -415,17 +423,6 @@ static int rcar_gen3_phy_usb2_init(struct phy *p)
 	struct rcar_gen3_chan *channel = rphy->ch;
 	void __iomem *usb2_base = channel->base;
 	u32 val;
-	int ret;
-
-	if (!rcar_gen3_is_any_rphy_initialized(channel) && channel->irq >= 0) {
-		INIT_WORK(&channel->work, rcar_gen3_phy_usb2_work);
-		ret = request_irq(channel->irq, rcar_gen3_phy_usb2_irq,
-				  IRQF_SHARED, dev_name(channel->dev), channel);
-		if (ret < 0) {
-			dev_err(channel->dev, "No irq handler (%d)\n", channel->irq);
-			return ret;
-		}
-	}
 
 	/* Initialize USB2 part */
 	val = readl(usb2_base + USB2_INT_ENABLE);
@@ -461,9 +458,6 @@ static int rcar_gen3_phy_usb2_exit(struct phy *p)
 		val &= ~USB2_INT_ENABLE_UCOM_INTEN;
 	writel(val, usb2_base + USB2_INT_ENABLE);
 
-	if (channel->irq >= 0 && !rcar_gen3_is_any_rphy_initialized(channel))
-		free_irq(channel->irq, channel);
-
 	return 0;
 }
 
@@ -612,7 +606,7 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev)
 	struct phy_provider *provider;
 	struct resource *res;
 	const struct phy_ops *phy_usb2_ops;
-	int ret = 0, i;
+	int ret = 0, i, irq;
 
 	if (!dev->of_node) {
 		dev_err(dev, "This driver needs device tree\n");
@@ -628,8 +622,6 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev)
 	if (IS_ERR(channel->base))
 		return PTR_ERR(channel->base);
 
-	/* get irq number here and request_irq for OTG in phy_init */
-	channel->irq = platform_get_irq_optional(pdev, 0);
 	channel->dr_mode = rcar_gen3_get_dr_mode(dev->of_node);
 	if (channel->dr_mode != USB_DR_MODE_UNKNOWN) {
 		channel->is_otg_channel = true;
@@ -683,6 +675,20 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, channel);
 	channel->dev = dev;
+
+	irq = platform_get_irq_optional(pdev, 0);
+	if (irq < 0 && irq != -ENXIO) {
+		ret = irq;
+		goto error;
+	} else if (irq > 0) {
+		INIT_WORK(&channel->work, rcar_gen3_phy_usb2_work);
+		ret = devm_request_irq(dev, irq, rcar_gen3_phy_usb2_irq,
+				       IRQF_SHARED, dev_name(dev), channel);
+		if (ret < 0) {
+			dev_err(dev, "Failed to request irq (%d)\n", irq);
+			goto error;
+		}
+	}
 
 	provider = devm_of_phy_provider_register(dev, rcar_gen3_phy_usb2_xlate);
 	if (IS_ERR(provider)) {
Signature · 1 line
2.53.0
author maintainer
#417 lore raw

PATCH 393/491 phy: renesas: rcar-gen3-usb2: Lock around hardware registers and driver data

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Claudiu Beznea <[email protected]>

commit 55a387ebb9219cbe4edfa8ba9996ccb0e7ad4932 upstream.

The phy-rcar-gen3-usb2 driver exposes four individual PHYs that are requested and configured by PHY users. The struct phy_ops APIs access the same set of registers to configure all PHYs. Additionally, PHY settings can be modified through sysfs or an IRQ handler. While some struct phy_ops APIs are protected by a driver-wide mutex, others rely on individual PHY-specific mutexes.

This approach can lead to various issues, including:
1/ the IRQ handler may interrupt PHY settings in progress, racing with
hardware configuration protected by a mutex lock
2/ due to msleep(20) in rcar_gen3_init_otg(), while a configuration thread
suspends to wait for the delay, another thread may try to configure another PHY (with phy_init() + phy_power_on()); re-running the phy_init() goes to the exact same configuration code, re-running the same hardware configuration on the same set of registers (and bits) which might impact the result of the msleep for the 1st configuring thread
3/ sysfs can configure the hardware (though role_store()) and it can
still race with the phy_init()/phy_power_on() APIs calling into the drivers struct phy_ops

To address these issues, add a spinlock to protect hardware register access and driver private data structures (e.g., calls to rcar_gen3_is_any_rphy_initialized()). Checking driver-specific data remains necessary as all PHY instances share common settings. With this change, the existing mutex protection is removed and the cleanup.h helpers are used.

While at it, to keep the code simpler, do not skip regulator_enable()/regulator_disable() APIs in rcar_gen3_phy_usb2_power_on()/rcar_gen3_phy_usb2_power_off() as the regulators enable/disable operations are reference counted anyway.

[claudiu.beznea:
- in rcar_gen3_init_otg(): fixed conflict by droppping ch->soc_no_adp_ctrl check
- in rcar_gen3_phy_usb2_irq() use spin_lock()/spin_unlock() as scoped_guard() is not avaialable in v5.10
- in probe(): replace mutex_init() with spin_lock_init()
- rcar_gen3_phy_usb2_power_off() replaced scoped_guard() as it is not available in v5.10
- in rcar_gen3_phy_usb2_power_on() droppped guard to avoid compilation warning "ISO C90 forbids mixed declarations and code"]

---
 drivers/phy/renesas/phy-rcar-gen3-usb2.c | 41 +++++++++++++++---------
 1 file changed, 26 insertions(+), 15 deletions(-)
drivers/phy/renesas/phy-rcar-gen3-usb2.c +26 -15
index f66e0daa23648..558c07512c05b 100644
@@ -9,6 +9,7 @@
  * Copyright (C) 2014 Cogent Embedded, Inc.
  */
 
+#include <linux/cleanup.h>
 #include <linux/extcon-provider.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
@@ -108,7 +109,7 @@ struct rcar_gen3_chan {
 	struct rcar_gen3_phy rphys[NUM_OF_PHYS];
 	struct regulator *vbus;
 	struct work_struct work;
-	struct mutex lock;	/* protects rphys[...].powered */
+	spinlock_t lock;	/* protects access to hardware and driver data structure. */
 	enum usb_dr_mode dr_mode;
 	bool extcon_host;
 	bool is_otg_channel;
@@ -317,6 +318,8 @@ static ssize_t role_store(struct device *dev, struct device_attribute *attr,
 	bool is_b_device;
 	enum phy_mode cur_mode, new_mode;
 
+	guard(spinlock_irqsave)(&ch->lock);
+
 	if (!ch->is_otg_channel || !rcar_gen3_is_any_otg_rphy_initialized(ch))
 		return -EIO;
 
@@ -404,6 +407,8 @@ static irqreturn_t rcar_gen3_phy_usb2_irq(int irq, void *_ch)
 	if (pm_runtime_suspended(dev))
 		goto rpm_put;
 
+	spin_lock(&ch->lock);
+
 	status = readl(usb2_base + USB2_OBINTSTA);
 	if (status & USB2_OBINT_BITS) {
 		dev_vdbg(dev, "%s: %08x\n", __func__, status);
@@ -412,6 +417,8 @@ static irqreturn_t rcar_gen3_phy_usb2_irq(int irq, void *_ch)
 		ret = IRQ_HANDLED;
 	}
 
+	spin_unlock(&ch->lock);
+
 rpm_put:
 	pm_runtime_put_noidle(dev);
 	return ret;
@@ -424,6 +431,8 @@ static int rcar_gen3_phy_usb2_init(struct phy *p)
 	void __iomem *usb2_base = channel->base;
 	u32 val;
 
+	guard(spinlock_irqsave)(&channel->lock);
+
 	/* Initialize USB2 part */
 	val = readl(usb2_base + USB2_INT_ENABLE);
 	val |= USB2_INT_ENABLE_UCOM_INTEN | rphy->int_enable_bits;
@@ -450,6 +459,8 @@ static int rcar_gen3_phy_usb2_exit(struct phy *p)
 	void __iomem *usb2_base = channel->base;
 	u32 val;
 
+	guard(spinlock_irqsave)(&channel->lock);
+
 	rphy->initialized = false;
 
 	val = readl(usb2_base + USB2_INT_ENABLE);
@@ -466,19 +477,21 @@ static int rcar_gen3_phy_usb2_power_on(struct phy *p)
 	struct rcar_gen3_phy *rphy = phy_get_drvdata(p);
 	struct rcar_gen3_chan *channel = rphy->ch;
 	void __iomem *usb2_base = channel->base;
+	unsigned long flags;
 	u32 val;
 	int ret = 0;
 
-	mutex_lock(&channel->lock);
-	if (!rcar_gen3_are_all_rphys_power_off(channel))
-		goto out;
-
 	if (channel->vbus) {
 		ret = regulator_enable(channel->vbus);
 		if (ret)
-			goto out;
+			return ret;
 	}
 
+	spin_lock_irqsave(&channel->lock, flags);
+
+	if (!rcar_gen3_are_all_rphys_power_off(channel))
+		goto out;
+
 	val = readl(usb2_base + USB2_USBCTR);
 	val |= USB2_USBCTR_PLL_RST;
 	writel(val, usb2_base + USB2_USBCTR);
@@ -488,7 +501,8 @@ static int rcar_gen3_phy_usb2_power_on(struct phy *p)
 out:
 	/* The powered flag should be set for any other phys anyway */
 	rphy->powered = true;
-	mutex_unlock(&channel->lock);
+
+	spin_unlock_irqrestore(&channel->lock, flags);
 
 	return 0;
 }
@@ -497,20 +511,16 @@ static int rcar_gen3_phy_usb2_power_off(struct phy *p)
 {
 	struct rcar_gen3_phy *rphy = phy_get_drvdata(p);
 	struct rcar_gen3_chan *channel = rphy->ch;
+	unsigned long flags;
 	int ret = 0;
 
-	mutex_lock(&channel->lock);
+	spin_lock_irqsave(&channel->lock, flags);
 	rphy->powered = false;
-
-	if (!rcar_gen3_are_all_rphys_power_off(channel))
-		goto out;
+	spin_unlock_irqrestore(&channel->lock, flags);
 
 	if (channel->vbus)
 		ret = regulator_disable(channel->vbus);
 
-out:
-	mutex_unlock(&channel->lock);
-
 	return ret;
 }
 
@@ -650,7 +660,8 @@ static int rcar_gen3_phy_usb2_probe(struct platform_device *pdev)
 		goto error;
 	}
 
-	mutex_init(&channel->lock);
+	spin_lock_init(&channel->lock);
+
 	for (i = 0; i < NUM_OF_PHYS; i++) {
 		channel->rphys[i].phy = devm_phy_create(dev, NULL,
 							phy_usb2_ops);
Signature · 1 line
2.53.0
#418 lore raw

Re: [PATCH 5.10 393/491] phy: renesas: rcar-gen3-usb2: Lock around hardware registers and driver data

Quoted 7 lines · Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Claudiu Beznea <[email protected]>

commit 55a387ebb9219cbe4edfa8ba9996ccb0e7ad4932 upstream.

[...]

This backport wrongly omitted the conversion of msleep() to mdelay() inside rcar_gen3_init_otg(), which is now called in atomic context.

Ben.

Signature · 2 lines
Ben Hutchings
Once a job is fouled up, anything done to improve it makes it worse.
#419 lore raw

PATCH 1/1 phy: renesas: rcar-gen3-usb2: Fix msleep() in atomic context

The backport of commit 55a387ebb921 "phy: renesas: rcar-gen3-usb2:
Lock around hardware registers and driver data" to 5.10 was not quite right. The upstream version converted an msleep() call to mdelay() in rcar_gen3_init_otg(), which now runs in atomic context, but this change was missing from the backport. Fix that.

Fixes 0f86a559900f ("phy: renesas: rcar-gen3-usb2: Lock around hardware registers and driver data")
---
 drivers/phy/renesas/phy-rcar-gen3-usb2.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/phy/renesas/phy-rcar-gen3-usb2.c +1 -1
index 5166a115879e..90f2a0e5b2aa 100644
@@ -386,7 +386,7 @@ static void rcar_gen3_init_otg(struct rcar_gen3_chan *ch)
 	val = readl(usb2_base + USB2_ADPCTRL);
 	writel(val | USB2_ADPCTRL_IDPULLUP, usb2_base + USB2_ADPCTRL);
 
-	msleep(20);
+	mdelay(20);
 
 	writel(0xffffffff, usb2_base + USB2_OBINTSTA);
 	writel(USB2_OBINT_BITS, usb2_base + USB2_OBINTEN);
author maintainer
#420 lore raw

PATCH 394/491 phy: renesas: rcar-gen3-usb2: Assert PLL reset on PHY power off

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Claudiu Beznea <[email protected]>

commit 9ce71e85b29eb63e48e294479742e670513f03a0 upstream.

Assert PLL reset on PHY power off. This saves power.

[claudiu.beznea: fixed conflict in rcar_gen3_phy_usb2_power_off() by
using spin_lock_irqsave()/spin_unlock_irqrestore() instead of scoped_guard()]

---
 drivers/phy/renesas/phy-rcar-gen3-usb2.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
drivers/phy/renesas/phy-rcar-gen3-usb2.c +8 -1
index 558c07512c05b..9fcbde094699d 100644
@@ -516,8 +516,15 @@ static int rcar_gen3_phy_usb2_power_off(struct phy *p)
 
 	spin_lock_irqsave(&channel->lock, flags);
 	rphy->powered = false;
-	spin_unlock_irqrestore(&channel->lock, flags);
 
+	if (rcar_gen3_are_all_rphys_power_off(channel)) {
+		u32 val = readl(channel->base + USB2_USBCTR);
+
+		val |= USB2_USBCTR_PLL_RST;
+		writel(val, channel->base + USB2_USBCTR);
+	}
+	spin_unlock_irqrestore(&channel->lock, flags);
+
 	if (channel->vbus)
 		ret = regulator_disable(channel->vbus);
 
Signature · 1 line
2.53.0
author maintainer
#421 lore raw

PATCH 395/491 bridge: br_nd_send: validate ND option lengths

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yang Yang <[email protected]>

commit 850837965af15707fd3142c1cf3c5bfaf022299b upstream.

br_nd_send() walks ND options according to option-provided lengths. A malformed option can make the parser advance beyond the computed option span or use a too-short source LLADDR option payload.

Validate option lengths against the remaining NS option area before advancing, and only read source LLADDR when the option is large enough for an Ethernet address.

---
 net/bridge/br_arp_nd_proxy.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
net/bridge/br_arp_nd_proxy.c +4 -2
@@ -285,12 +285,14 @@ static void br_nd_send(struct net_bridge
 	ns_olen = request->len - (skb_network_offset(request) +
 				  sizeof(struct ipv6hdr)) - sizeof(*ns);
 	for (i = 0; i < ns_olen - 1; i += (ns->opt[i + 1] << 3)) {
-		if (!ns->opt[i + 1]) {
+		if (!ns->opt[i + 1] || i + (ns->opt[i + 1] << 3) > ns_olen) {
 			kfree_skb(reply);
 			return;
 		}
 		if (ns->opt[i] == ND_OPT_SOURCE_LL_ADDR) {
-			daddr = ns->opt + i + sizeof(struct nd_opt_hdr);
+			if ((ns->opt[i + 1] << 3) >=
+			    sizeof(struct nd_opt_hdr) + ETH_ALEN)
+				daddr = ns->opt + i + sizeof(struct nd_opt_hdr);
 			break;
 		}
 	}
author maintainer
#422 lore raw

PATCH 396/491 cdc-acm: new quirk for EPSON HMD

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Oliver Neukum <[email protected]>

commit f97e96c303d689708f7f713d8f3afcc31f1237e9 upstream.

This device has a union descriptor that is just garbage and needs a custom descriptor.
In principle this could be done with a (conditionally activated) heuristic. That would match more devices without a need for defining a new quirk. However, this always carries the risk that the heuristics does the wrong thing and leads to more breakage. Defining the quirk and telling it exactly what to do is the safe and conservative approach.

---
 drivers/usb/class/cdc-acm.c |    9 +++++++++
 drivers/usb/class/cdc-acm.h |    1 +
 2 files changed, 10 insertions(+)
drivers/usb/class/cdc-acm.c +9 -0
@@ -1218,6 +1218,12 @@ static int acm_probe(struct usb_interfac
 		if (!data_interface || !control_interface)
 			return -ENODEV;
 		goto skip_normal_probe;
+	} else if (quirks == NO_UNION_12) {
+		data_interface = usb_ifnum_to_if(usb_dev, 2);
+		control_interface = usb_ifnum_to_if(usb_dev, 1);
+		if (!data_interface || !control_interface)
+			 return -ENODEV;
+		goto skip_normal_probe;
 	}
 
 	/* normal probing*/
@@ -1749,6 +1755,9 @@ static const struct usb_device_id acm_id
 	{ USB_DEVICE(0x045b, 0x024D),	/* Renesas R-Car E3 USB Download mode */
 	.driver_info = DISABLE_ECHO,	/* Don't echo banner */
 	},
+	{ USB_DEVICE(0x04b8, 0x0d12),	/* EPSON HMD Com&Sens */
+	.driver_info = NO_UNION_12,	/* union descriptor is garbage */
+	},
 	{ USB_DEVICE(0x0e8d, 0x0003), /* FIREFLY, MediaTek Inc; [email protected] */
 	.driver_info = NO_UNION_NORMAL, /* has no union descriptor */
 	},
drivers/usb/class/cdc-acm.h +1 -0
@@ -142,3 +142,4 @@ struct acm {
 #define SEND_ZERO_PACKET		BIT(6)
 #define DISABLE_ECHO			BIT(7)
 #define MISSING_CAP_BRK			BIT(8)
+#define NO_UNION_12			BIT(9)
author maintainer
#423 lore raw

PATCH 397/491 comedi: dt2815: add hardware detection to prevent crash

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Deepanshu Kartikey <[email protected]>

commit 93853512f565e625df2397f0d8050d6aafd7c3ad upstream.

The dt2815 driver crashes when attached to I/O ports without actual hardware present. This occurs because syzkaller or users can attach the driver to arbitrary I/O addresses via COMEDI_DEVCONFIG ioctl.

When no hardware exists at the specified port, inb() operations return 0xff (floating bus), but outb() operations can trigger page faults due to undefined behavior, especially under race conditions:

  BUG: unable to handle page fault for address: 000000007fffff90
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  RIP: 0010:dt2815_attach+0x6e0/0x1110

Add hardware detection by reading the status register before attempting any write operations. If the read returns 0xff, assume no hardware is present and fail the attach with -ENODEV. This prevents crashes from outb() operations on non-existent hardware.

---
 drivers/staging/comedi/drivers/dt2815.c |   12 ++++++++++++
 1 file changed, 12 insertions(+)
drivers/staging/comedi/drivers/dt2815.c +12 -0
@@ -176,6 +176,18 @@ static int dt2815_attach(struct comedi_d
 		    ? current_range_type : voltage_range_type;
 	}
 
+	/*
+	 * Check if hardware is present before attempting any I/O operations.
+	 * Reading 0xff from status register typically indicates no hardware
+	 * on the bus (floating bus reads as all 1s).
+	 */
+	if (inb(dev->iobase + DT2815_STATUS) == 0xff) {
+		dev_err(dev->class_dev,
+			"No hardware detected at I/O base 0x%lx\n",
+			dev->iobase);
+		return -ENODEV;
+	}
+
 	/* Init the 2815 */
 	outb(0x00, dev->iobase + DT2815_STATUS);
 	for (i = 0; i < 100; i++) {
author maintainer
#424 lore raw

PATCH 398/491 comedi: Reinit dev->spinlock between attachments to low-level drivers

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ian Abbott <[email protected]>

commit 4b9a9a6d71e3e252032f959fb3895a33acb5865c upstream.

struct comedi_device is the main controlling structure for a COMEDI device created by the COMEDI subsystem. It contains a member spinlock containing a spin-lock that is initialized by the COMEDI subsystem, but is reserved for use by a low-level driver attached to the COMEDI device (at least since commit 25436dc9d84f ("Staging: comedi: remove RT code")).

Some COMEDI devices (those created on initialization of the COMEDI subsystem when the "comedi.comedi_num_legacy_minors" parameter is non-zero) can be attached to different low-level drivers over their lifetime using the COMEDI_DEVCONFIG ioctl command. This can result in inconsistent lock states being reported when there is a mismatch in the spin-lock locking levels used by each low-level driver to which the COMEDI device has been attached. Fix it by reinitializing dev->spinlock before calling the low-level driver's attach function pointer if CONFIG_LOCKDEP is enabled.

---
 drivers/staging/comedi/drivers.c |    8 ++++++++
 1 file changed, 8 insertions(+)
drivers/staging/comedi/drivers.c +8 -0
@@ -1000,6 +1000,14 @@ int comedi_device_attach(struct comedi_d
 		ret = -EIO;
 		goto out;
 	}
+	if (IS_ENABLED(CONFIG_LOCKDEP)) {
+		/*
+		 * dev->spinlock is for private use by the attached low-level
+		 * driver.  Reinitialize it to stop lock-dependency tracking
+		 * between attachments to different low-level drivers.
+		 */
+		spin_lock_init(&dev->spinlock);
+	}
 	dev->driver = driv;
 	dev->board_name = dev->board_ptr ? *(const char **)dev->board_ptr
 					 : dev->driver->driver_name;
author maintainer
#425 lore raw

PATCH 399/491 comedi: ni_atmio16d: Fix invalid clean-up after failed attach

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ian Abbott <[email protected]>

commit 101ab946b79ad83b36d5cfd47de587492a80acf0 upstream.

If the driver's COMEDI "attach" handler function (atmio16d_attach()) returns an error, the COMEDI core will call the driver's "detach" handler function (atmio16d_detach()) to clean up. This calls reset_atmio16d() unconditionally, but depending on where the error occurred in the attach handler, the device may not have been sufficiently initialized to call reset_atmio16d(). It uses dev->iobase as the I/O port base address and dev->private as the pointer to the COMEDI device's private data structure. dev->iobase may still be set to its initial value of 0, which would result in undesired writes to low I/O port addresses. dev->private may still be NULL, which would result in null pointer dereferences.

Fix atmio16d_detach() by checking that dev->private is valid (non-null) before calling reset_atmio16d(). This implies that dev->iobase was set correctly since that is set up before dev->private.

Fixes 2323b276308a ("Staging: comedi: add ni_at_atmio16d driver")
Cc stable <[email protected]>
---
 drivers/staging/comedi/drivers/ni_atmio16d.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/staging/comedi/drivers/ni_atmio16d.c +2 -1
@@ -699,7 +699,8 @@ static int atmio16d_attach(struct comedi
 
 static void atmio16d_detach(struct comedi_device *dev)
 {
-	reset_atmio16d(dev);
+	if (dev->private)
+		reset_atmio16d(dev);
 	comedi_legacy_detach(dev);
 }
 
author maintainer
#426 lore raw

PATCH 400/491 comedi: me_daq: Fix potential overrun of firmware buffer

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ian Abbott <[email protected]>

commit cc797d4821c754c701d9714b58bea947e31dbbe0 upstream.

me2600_xilinx_download() loads the firmware that was requested by request_firmware(). It is possible for it to overrun the source buffer because it blindly trusts the file format. It reads a data stream length from the first 4 bytes into variable file_length and reads the data stream contents of length file_length from offset 16 onwards. Although it checks that the supplied firmware is at least 16 bytes long, it does not check that it is long enough to contain the data stream.

Add a test to ensure that the supplied firmware is long enough to contain the header and the data stream. On failure, log an error and return -EINVAL.

Fixes 85acac61096f9 ("Staging: comedi: add me_daq driver")
Cc stable <[email protected]>
---
 drivers/staging/comedi/drivers/me_daq.c |   35 +++++++++++++++++---------------
 1 file changed, 19 insertions(+), 16 deletions(-)
drivers/staging/comedi/drivers/me_daq.c +19 -16
@@ -345,6 +345,25 @@ static int me2600_xilinx_download(struct
 	unsigned int file_length;
 	unsigned int i;
 
+	/*
+	 * Format of the firmware
+	 * Build longs from the byte-wise coded header
+	 * Byte 1-3:   length of the array
+	 * Byte 4-7:   version
+	 * Byte 8-11:  date
+	 * Byte 12-15: reserved
+	 */
+	if (size >= 4) {
+		file_length = (((unsigned int)data[0] & 0xff) << 24) +
+			      (((unsigned int)data[1] & 0xff) << 16) +
+			      (((unsigned int)data[2] & 0xff) << 8) +
+			      ((unsigned int)data[3] & 0xff);
+	}
+	if (size < 16 || file_length > size - 16) {
+		dev_err(dev->class_dev, "Firmware length inconsistency\n");
+		return -EINVAL;
+	}
+
 	/* disable irq's on PLX */
 	writel(0x00, devpriv->plx_regbase + PLX9052_INTCSR);
 
@@ -359,22 +378,6 @@ static int me2600_xilinx_download(struct
 	sleep(1);
 
 	/*
-	 * Format of the firmware
-	 * Build longs from the byte-wise coded header
-	 * Byte 1-3:   length of the array
-	 * Byte 4-7:   version
-	 * Byte 8-11:  date
-	 * Byte 12-15: reserved
-	 */
-	if (size < 16)
-		return -EINVAL;
-
-	file_length = (((unsigned int)data[0] & 0xff) << 24) +
-	    (((unsigned int)data[1] & 0xff) << 16) +
-	    (((unsigned int)data[2] & 0xff) << 8) +
-	    ((unsigned int)data[3] & 0xff);
-
-	/*
 	 * Loop for writing firmware byte by byte to xilinx
 	 * Firmware data start at offset 16
 	 */
author maintainer
#427 lore raw

PATCH 401/491 comedi: me4000: Fix potential overrun of firmware buffer

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ian Abbott <[email protected]>

commit 3fb43a7a5b44713f892c58ead2e5f3a1bc9f4ee7 upstream.

me4000_xilinx_download() loads the firmware that was requested by request_firmware(). It is possible for it to overrun the source buffer because it blindly trusts the file format. It reads a data stream length from the first 4 bytes into variable file_length and reads the data stream contents of length file_length from offset 16 onwards.

Add a test to ensure that the supplied firmware is long enough to contain the header and the data stream. On failure, log an error and return -EINVAL.

Note: The firmware loading was totally broken before commit ac584af59945 ("staging: comedi: me4000: fix firmware downloading"), but that is the most sensible target for this fix.

Fixes ac584af59945 ("staging: comedi: me4000: fix firmware downloading")
Cc stable <[email protected]>
---
 drivers/staging/comedi/drivers/me4000.c |   16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)
drivers/staging/comedi/drivers/me4000.c +12 -4
@@ -316,6 +316,18 @@ static int me4000_xilinx_download(struct
 	unsigned int val;
 	unsigned int i;
 
+	/* Get data stream length from header. */
+	if (size >= 4) {
+		file_length = (((unsigned int)data[0] & 0xff) << 24) +
+			      (((unsigned int)data[1] & 0xff) << 16) +
+			      (((unsigned int)data[2] & 0xff) << 8) +
+			      ((unsigned int)data[3] & 0xff);
+	}
+	if (size < 16 || file_length > size - 16) {
+		dev_err(dev->class_dev, "Firmware length inconsistency\n");
+		return -EINVAL;
+	}
+
 	if (!xilinx_iobase)
 		return -ENODEV;
 
@@ -347,10 +359,6 @@ static int me4000_xilinx_download(struct
 	outl(val, devpriv->plx_regbase + PLX9052_CNTRL);
 
 	/* Download Xilinx firmware */
-	file_length = (((unsigned int)data[0] & 0xff) << 24) +
-		      (((unsigned int)data[1] & 0xff) << 16) +
-		      (((unsigned int)data[2] & 0xff) << 8) +
-		      ((unsigned int)data[3] & 0xff);
 	usleep_range(10, 1000);
 
 	for (i = 0; i < file_length; i++) {
author maintainer
#428 lore raw

PATCH 402/491 netfilter: ipset: drop logically empty buckets in mtype_del

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yifan Wu <[email protected]>

commit 9862ef9ab0a116c6dca98842aab7de13a252ae02 upstream.

mtype_del() counts empty slots below n->pos in k, but it only drops the bucket when both n->pos and k are zero. This misses buckets whose live entries have all been removed while n->pos still points past deleted slots.

Treat a bucket as empty when all positions below n->pos are unused and release it directly instead of shrinking it further.

---
 net/netfilter/ipset/ip_set_hash_gen.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/netfilter/ipset/ip_set_hash_gen.h +1 -1
@@ -1108,7 +1108,7 @@ mtype_del(struct ip_set *set, void *valu
 			if (!test_bit(i, n->used))
 				k++;
 		}
-		if (n->pos == 0 && k == 0) {
+		if (k == n->pos) {
 			t->hregion[r].ext_size -= ext_size(n->size, dsize);
 			rcu_assign_pointer(hbucket(t, key), NULL);
 			kfree_rcu(n, rcu);
author maintainer
#429 lore raw

PATCH 403/491 vxlan: validate ND option lengths in vxlan_na_create

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yang Yang <[email protected]>

commit afa9a05e6c4971bd5586f1b304e14d61fb3d9385 upstream.

vxlan_na_create() walks ND options according to option-provided lengths. A malformed option can make the parser advance beyond the computed option span or use a too-short source LLADDR option payload.

Validate option lengths against the remaining NS option area before advancing, and only read source LLADDR when the option is large enough for an Ethernet address.

---
 drivers/net/vxlan/vxlan_core.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/net/vxlan/vxlan_core.c +4 -2
@@ -2098,12 +2098,14 @@ static struct sk_buff *vxlan_na_create(s
 	ns_olen = request->len - skb_network_offset(request) -
 		sizeof(struct ipv6hdr) - sizeof(*ns);
 	for (i = 0; i < ns_olen-1; i += (ns->opt[i+1]<<3)) {
-		if (!ns->opt[i + 1]) {
+		if (!ns->opt[i + 1] || i + (ns->opt[i + 1] << 3) > ns_olen) {
 			kfree_skb(reply);
 			return NULL;
 		}
 		if (ns->opt[i] == ND_OPT_SOURCE_LL_ADDR) {
-			daddr = ns->opt + i + sizeof(struct nd_opt_hdr);
+			if ((ns->opt[i + 1] << 3) >=
+			    sizeof(struct nd_opt_hdr) + ETH_ALEN)
+				daddr = ns->opt + i + sizeof(struct nd_opt_hdr);
 			break;
 		}
 	}
author maintainer
#430 lore raw

PATCH 404/491 net: ftgmac100: fix ring allocation unwind on open failure

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yufan Chen <[email protected]>

commit c0fd0fe745f5e8c568d898cd1513d0083e46204a upstream.

ftgmac100_alloc_rings() allocates rx_skbs, tx_skbs, rxdes, txdes, and rx_scratch in stages. On intermediate failures it returned -ENOMEM directly, leaking resources allocated earlier in the function.

Rework the failure path to use staged local unwind labels and free allocated resources in reverse order before returning -ENOMEM. This matches common netdev allocation cleanup style.

Fixes d72e01a0430f ("ftgmac100: Use a scratch buffer for failed RX allocations")
---
 drivers/net/ethernet/faraday/ftgmac100.c |   28 ++++++++++++++++++++++++----
 1 file changed, 24 insertions(+), 4 deletions(-)
drivers/net/ethernet/faraday/ftgmac100.c +24 -4
@@ -931,19 +931,19 @@ static int ftgmac100_alloc_rings(struct
 	priv->tx_skbs = kcalloc(MAX_TX_QUEUE_ENTRIES, sizeof(void *),
 				GFP_KERNEL);
 	if (!priv->tx_skbs)
-		return -ENOMEM;
+		goto err_free_rx_skbs;
 
 	/* Allocate descriptors */
 	priv->rxdes = dma_alloc_coherent(priv->dev,
 					 MAX_RX_QUEUE_ENTRIES * sizeof(struct ftgmac100_rxdes),
 					 &priv->rxdes_dma, GFP_KERNEL);
 	if (!priv->rxdes)
-		return -ENOMEM;
+		goto err_free_tx_skbs;
 	priv->txdes = dma_alloc_coherent(priv->dev,
 					 MAX_TX_QUEUE_ENTRIES * sizeof(struct ftgmac100_txdes),
 					 &priv->txdes_dma, GFP_KERNEL);
 	if (!priv->txdes)
-		return -ENOMEM;
+		goto err_free_rxdes;
 
 	/* Allocate scratch packet buffer */
 	priv->rx_scratch = dma_alloc_coherent(priv->dev,
@@ -951,9 +951,29 @@ static int ftgmac100_alloc_rings(struct
 					      &priv->rx_scratch_dma,
 					      GFP_KERNEL);
 	if (!priv->rx_scratch)
-		return -ENOMEM;
+		goto err_free_txdes;
 
 	return 0;
+
+err_free_txdes:
+	dma_free_coherent(priv->dev,
+			  MAX_TX_QUEUE_ENTRIES *
+			  sizeof(struct ftgmac100_txdes),
+			  priv->txdes, priv->txdes_dma);
+	priv->txdes = NULL;
+err_free_rxdes:
+	dma_free_coherent(priv->dev,
+			  MAX_RX_QUEUE_ENTRIES *
+			  sizeof(struct ftgmac100_rxdes),
+			  priv->rxdes, priv->rxdes_dma);
+	priv->rxdes = NULL;
+err_free_tx_skbs:
+	kfree(priv->tx_skbs);
+	priv->tx_skbs = NULL;
+err_free_rx_skbs:
+	kfree(priv->rx_skbs);
+	priv->rx_skbs = NULL;
+	return -ENOMEM;
 }
 
 static void ftgmac100_init_rings(struct ftgmac100 *priv)
author maintainer
#431 lore raw

PATCH 405/491 thunderbolt: Fix property read in nhi_wake_supported()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Konrad Dybcio <[email protected]>

commit 73a505dc48144ec72e25874e2b2a72487b02d3bc upstream.

device_property_read_foo() returns 0 on success and only then modifies 'val'. Currently, val is left uninitialized if the aforementioned function returns non-zero, making nhi_wake_supported() return true almost always (random != 0) if the property is not present in device firmware.

Invert the check to make it make sense.

Fixes 3cdb9446a117 ("thunderbolt: Add support for Intel Ice Lake")
---
 drivers/thunderbolt/nhi.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/thunderbolt/nhi.c +1 -1
@@ -899,7 +899,7 @@ static bool nhi_wake_supported(struct pc
 	 * If power rails are sustainable for wakeup from S4 this
 	 * property is set by the BIOS.
 	 */
-	if (device_property_read_u8(&pdev->dev, "WAKE_SUPPORTED", &val))
+	if (!device_property_read_u8(&pdev->dev, "WAKE_SUPPORTED", &val))
 		return !!val;
 
 	return true;
author maintainer
#432 lore raw

PATCH 406/491 USB: dummy-hcd: Fix locking/synchronization error

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alan Stern <[email protected]>

commit 616a63ff495df12863692ab3f9f7b84e3fa7a66d upstream.

Syzbot testing was able to provoke an addressing exception and crash in the usb_gadget_udc_reset() routine in drivers/usb/gadgets/udc/core.c, resulting from the fact that the routine was called with a second ("driver") argument of NULL. The bad caller was set_link_state() in dummy_hcd.c, and the problem arose because of a race between a USB reset and driver unbind.

These sorts of races were not supposed to be possible; commit 7dbd8f4cabd9 ("USB: dummy-hcd: Fix erroneous synchronization change"), along with a few followup commits, was written specifically to prevent them. As it turns out, there are (at least) two errors remaining in the code. Another patch will address the second error; this one is concerned with the first.

The error responsible for the syzbot crash occurred because the stop_activity() routine will sometimes drop and then re-acquire the dum->lock spinlock. A call to stop_activity() occurs in set_link_state() when handling an emulated USB reset, after the test of dum->ints_enabled and before the increment of dum->callback_usage. This allowed another thread (doing a driver unbind) to sneak in and grab the spinlock, and then clear dum->ints_enabled and dum->driver. Normally this other thread would have to wait for dum->callback_usage to go down to 0 before it would clear dum->driver, but in this case it didn't have to wait since dum->callback_usage had not yet been incremented.

The fix is to increment dum->callback_usage _before_ calling stop_activity() instead of after. Then the thread doing the unbind will not clear dum->driver until after the call to usb_gadget_udc_reset() safely returns and dum->callback_usage has been decremented again.

---
 drivers/usb/gadget/udc/dummy_hcd.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/usb/gadget/udc/dummy_hcd.c +6 -1
@@ -458,8 +458,13 @@ static void set_link_state(struct dummy_
 
 		/* Report reset and disconnect events to the driver */
 		if (dum->ints_enabled && (disconnect || reset)) {
-			stop_activity(dum);
 			++dum->callback_usage;
+			/*
+			 * stop_activity() can drop dum->lock, so it must
+			 * not come between the dum->ints_enabled test
+			 * and the ++dum->callback_usage.
+			 */
+			stop_activity(dum);
 			spin_unlock(&dum->lock);
 			if (reset)
 				usb_gadget_udc_reset(&dum->gadget, dum->driver);
author maintainer
#433 lore raw

PATCH 407/491 usb: gadget: dummy_hcd: fix premature URB completion when ZLP follows partial transfer

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sebastian Urban <[email protected]>

commit f50200dd44125e445a6164e88c217472fa79cdbc upstream.

When a gadget request is only partially transferred in transfer() because the per-frame bandwidth budget is exhausted, the loop advances to the next queued request. If that next request is a zero-length packet (ZLP), len evaluates to zero and the code takes the unlikely(len == 0) path, which sets is_short = 1. This bypasses the bandwidth guard ("limit < ep->ep.maxpacket && limit < len") that lives in the else branch and would otherwise break out of the loop for non-zero requests. The is_short path then completes the URB before all data from the first request has been transferred.

Reproducer (bulk IN, high speed):

Device side (FunctionFS with Linux AIO):
1. Queue a 65024-byte write via io_submit (127 * 512, i.e. a multiple of the HS bulk max packet size).
2. Immediately queue a zero-length write (ZLP) via io_submit.

Host side:
3. Submit a 65536-byte bulk IN URB.

Expected: URB completes with actual_length = 65024.
Actual: URB completes with actual_length = 53248, losing 11776
bytes that leak into subsequent URBs.

At high speed the per-frame budget is 53248 bytes (512 * 13 * 8). The 65024-byte request exhausts this budget after 53248 bytes, leaving the request incomplete (req->req.actual < req->req.length). Neither the request nor the URB is finished, and rescan is 0, so the loop advances to the ZLP. For the ZLP, dev_len = 0, so len = min(12288, 0) = 0, taking the unlikely(len == 0) path and setting is_short = 1. The is_short handler then sets *status = 0, completing the URB with only 53248 of the expected 65024 bytes.

Fix this by breaking out of the loop when the current request has remaining data (req->req.actual < req->req.length). The request resumes on the next timer tick, preserving correct data ordering.

---
 drivers/usb/gadget/udc/dummy_hcd.c |    6 ++++++
 1 file changed, 6 insertions(+)
drivers/usb/gadget/udc/dummy_hcd.c +6 -0
@@ -1521,6 +1521,12 @@ top:
 		/* rescan to continue with any other queued i/o */
 		if (rescan)
 			goto top;
+
+		/* request not fully transferred; stop iterating to
+		 * preserve data ordering across queued requests.
+		 */
+		if (req->req.actual < req->req.length)
+			break;
 	}
 	return sent;
 }
author maintainer
#434 lore raw

PATCH 408/491 nvmet-tcp: fix use-before-check of sg in bounds validation

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cengiz Can <[email protected]>

The stable backport of commit 52a0a9854934 ("nvmet-tcp: add bounds checks in nvmet_tcp_build_pdu_iovec") placed the bounds checks after the iov_len calculation:

    while (length) {
        u32 iov_len = min_t(u32, length, sg->length - sg_offset);

        if (!sg_remaining) {    /* too late: sg already dereferenced */

In mainline, the checks come first because C99 allows mid-block variable declarations. The stable backport moved the declaration to the top of the loop to satisfy C89 declaration rules, but this ended up placing the sg->length dereference before the sg_remaining and sg->length guards.

If sg_next() returns NULL at the end of the scatterlist, the next iteration dereferences a NULL pointer in the iov_len calculation before the sg_remaining check can prevent it.

Fix this by moving the iov_len declaration to function scope and keeping the assignment after the bounds checks, matching the ordering in mainline.

Fixes 043b4307a99f ("nvmet-tcp: add bounds checks in nvmet_tcp_build_pdu_iovec")
Cc YunJe Shin <[email protected]>
Cc Sagi Grimberg <[email protected]>
Cc Keith Busch <[email protected]>
---
 drivers/nvme/target/tcp.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/nvme/target/tcp.c +3 -3
index 5d8e57e5fdb18..6db9dcdbb3c34 100644
@@ -300,7 +300,7 @@ static void nvmet_tcp_build_pdu_iovec(struct nvmet_tcp_cmd *cmd)
 {
 	struct bio_vec *iov = cmd->iov;
 	struct scatterlist *sg;
-	u32 length, offset, sg_offset;
+	u32 length, offset, sg_offset, iov_len;
 	unsigned int sg_remaining;
 	int nr_pages;
 
@@ -317,8 +317,6 @@ static void nvmet_tcp_build_pdu_iovec(struct nvmet_tcp_cmd *cmd)
 	sg_remaining = cmd->req.sg_cnt - cmd->sg_idx;
 
 	while (length) {
-		u32 iov_len = min_t(u32, length, sg->length - sg_offset);
-
 		if (!sg_remaining) {
 			nvmet_tcp_fatal_error(cmd->queue);
 			return;
@@ -328,6 +326,8 @@ static void nvmet_tcp_build_pdu_iovec(struct nvmet_tcp_cmd *cmd)
 			return;
 		}
 
+		iov_len = min_t(u32, length, sg->length - sg_offset);
+
 		iov->bv_page = sg_page(sg);
 		iov->bv_len = iov_len;
 		iov->bv_offset = sg->offset + sg_offset;
Signature · 1 line
2.53.0
author maintainer
#435 lore raw

PATCH 409/491 usb: gadget: f_subset: Fix unbalanced refcnt in geth_free

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuen-Han Tsai <[email protected]>

commit caa27923aacd8a5869207842f2ab1657c6c0c7bc upstream.

geth_alloc() increments the reference count, but geth_free() fails to decrement it. This prevents the configuration of attributes via configfs after unlinking the function.

Decrement the reference count in geth_free() to ensure proper cleanup.

Fixes 02832e56f88a ("usb: gadget: f_subset: add configfs support")
---
 drivers/usb/gadget/function/f_subset.c |    7 +++++++
 1 file changed, 7 insertions(+)
drivers/usb/gadget/function/f_subset.c +7 -0
@@ -6,6 +6,7 @@
  * Copyright (C) 2008 Nokia Corporation
  */
 
+#include <linux/cleanup.h>
 #include <linux/slab.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
@@ -451,8 +452,14 @@ static struct usb_function_instance *get
 static void geth_free(struct usb_function *f)
 {
 	struct f_gether *eth;
+	struct f_gether_opts *opts;
+
+	opts = container_of(f->fi, struct f_gether_opts, func_inst);
 
 	eth = func_to_geth(f);
+	mutex_lock(&opts->lock);
+	opts->refcnt--;
+	mutex_unlock(&opts->lock);
 	kfree(eth);
 }
 
author maintainer
#436 lore raw

PATCH 410/491 usb: gadget: f_rndis: Protect RNDIS options with mutex

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuen-Han Tsai <[email protected]>

commit 8d8c68b1fc06ece60cf43e1306ff0f4ac121547e upstream.

The class/subclass/protocol options are suspectible to race conditions as they can be accessed concurrently through configfs.

Use existing mutex to protect these options. This issue was identified during code inspection.

Fixes 73517cf49bd4 ("usb: gadget: add RNDIS configfs options for class/subclass/protocol")
---
 drivers/usb/gadget/function/f_rndis.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/usb/gadget/function/f_rndis.c +3 -0
@@ -11,6 +11,7 @@
 
 /* #define VERBOSE_DEBUG */
 
+#include <linux/cleanup.h>
 #include <linux/slab.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
@@ -690,9 +691,11 @@ rndis_bind(struct usb_configuration *c,
 		f->os_desc_table[0].os_desc = &rndis_opts->rndis_os_desc;
 	}
 
+	mutex_lock(&rndis_opts->lock);
 	rndis_iad_descriptor.bFunctionClass = rndis_opts->class;
 	rndis_iad_descriptor.bFunctionSubClass = rndis_opts->subclass;
 	rndis_iad_descriptor.bFunctionProtocol = rndis_opts->protocol;
+	mutex_unlock(&rndis_opts->lock);
 
 	/*
 	 * in drivers/usb/gadget/configfs.c:configfs_composite_bind()
author maintainer
#437 lore raw

PATCH 411/491 usb: gadget: f_uac1_legacy: validate control request size

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Taegu Ha <[email protected]>

commit 6e0e34d85cd46ceb37d16054e97a373a32770f6c upstream.

f_audio_complete() copies req->length bytes into a 4-byte stack variable:

  u32 data = 0;
  memcpy(&data, req->buf, req->length);

req->length is derived from the host-controlled USB request path, which can lead to a stack out-of-bounds write.

Validate req->actual against the expected payload size for the supported control selectors and decode only the expected amount of data.

This avoids copying a host-influenced length into a fixed-size stack object.

---
 drivers/usb/gadget/function/f_uac1_legacy.c |   47 ++++++++++++++++++++++------
 1 file changed, 37 insertions(+), 10 deletions(-)
drivers/usb/gadget/function/f_uac1_legacy.c +37 -10
@@ -360,19 +360,46 @@ static int f_audio_out_ep_complete(struc
 static void f_audio_complete(struct usb_ep *ep, struct usb_request *req)
 {
 	struct f_audio *audio = req->context;
-	int status = req->status;
-	u32 data = 0;
 	struct usb_ep *out_ep = audio->out_ep;
 
-	switch (status) {
-
-	case 0:				/* normal completion? */
-		if (ep == out_ep)
+	switch (req->status) {
+	case 0:
+		if (ep == out_ep) {
 			f_audio_out_ep_complete(ep, req);
-		else if (audio->set_con) {
-			memcpy(&data, req->buf, req->length);
-			audio->set_con->set(audio->set_con, audio->set_cmd,
-					le16_to_cpu(data));
+		} else if (audio->set_con) {
+			struct usb_audio_control *con = audio->set_con;
+			u8 type = con->type;
+			u32 data;
+			bool valid_request = false;
+
+			switch (type) {
+			case UAC_FU_MUTE: {
+				u8 value;
+
+				if (req->actual == sizeof(value)) {
+					memcpy(&value, req->buf, sizeof(value));
+					data = value;
+					valid_request = true;
+				}
+				break;
+			}
+			case UAC_FU_VOLUME: {
+				__le16 value;
+
+				if (req->actual == sizeof(value)) {
+					memcpy(&value, req->buf, sizeof(value));
+					data = le16_to_cpu(value);
+					valid_request = true;
+				}
+				break;
+			}
+			}
+
+			if (valid_request)
+				con->set(con, audio->set_cmd, data);
+			else
+				usb_ep_set_halt(ep);
+
 			audio->set_con = NULL;
 		}
 		break;
author maintainer
#438 lore raw

PATCH 412/491 io_uring/tctx: work around xa_store() allocation error issue

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jens Axboe <[email protected]>

commit 7eb75ce7527129d7f1fee6951566af409a37a1c4 upstream.

syzbot triggered the following WARN_ON:

WARNING: CPU: 0 PID: 16 at io_uring/tctx.c:51 __io_uring_free+0xfa/0x140 io_uring/tctx.c:51

which is the

WARN_ON_ONCE(!xa_empty(&tctx->xa));

sanity check in __io_uring_free() when a io_uring_task is going through its final put. The syzbot test case includes injecting memory allocation failures, and it very much looks like xa_store() can fail one of its memory allocations and end up with ->head being non-NULL even though no entries exist in the xarray.

Until this issue gets sorted out, work around it by attempting to iterate entries in our xarray, and WARN_ON_ONCE() if one is found.

Reported-by [email protected]
[ Modify the function in io_uring.c because it's located here in v5.15. ]
---
 io_uring/io_uring.c |   13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)
io_uring/io_uring.c +12 -1
@@ -8524,8 +8524,19 @@ static int io_uring_alloc_task_context(s
 void __io_uring_free(struct task_struct *tsk)
 {
 	struct io_uring_task *tctx = tsk->io_uring;
+	struct io_tctx_node *node;
+	unsigned long index;
 
-	WARN_ON_ONCE(!xa_empty(&tctx->xa));
+	/*
+	 * Fault injection forcing allocation errors in the xa_store() path
+	 * can lead to xa_empty() returning false, even though no actual
+	 * node is stored in the xarray. Until that gets sorted out, attempt
+	 * an iteration here and warn if any entries are found.
+	 */
+	xa_for_each(&tctx->xa, index, node) {
+		WARN_ON_ONCE(1);
+		break;
+	}
 	WARN_ON_ONCE(tctx->io_wq);
 	WARN_ON_ONCE(tctx->cached_refs);
 
author maintainer
#439 lore raw

PATCH 413/491 lib/crypto: chacha: Zeroize permuted_state before it leaves scope

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Biggers <[email protected]>

commit e5046823f8fa3677341b541a25af2fcb99a5b1e0 upstream.

Since the ChaCha permutation is invertible, the local variable 'permuted_state' is sufficient to compute the original 'state', and thus the key, even after the permutation has been done.

While the kernel is quite inconsistent about zeroizing secrets on the stack (and some prominent userspace crypto libraries don't bother at all since it's not guaranteed to work anyway), the kernel does try to do it as a best practice, especially in cases involving the RNG.

Thus, explicitly zeroize 'permuted_state' before it goes out of scope.

Fixes c08d0e647305 ("crypto: chacha20 - Add a generic ChaCha20 stream cipher implementation")
---
 lib/crypto/chacha.c |    4 ++++
 1 file changed, 4 insertions(+)
lib/crypto/chacha.c +4 -0
@@ -86,6 +86,8 @@ void chacha_block_generic(u32 *state, u8
 		put_unaligned_le32(x[i] + state[i], &stream[i * sizeof(u32)]);
 
 	state[12]++;
+
+	memzero_explicit(x, sizeof(x));
 }
 EXPORT_SYMBOL(chacha_block_generic);
 
@@ -110,5 +112,7 @@ void hchacha_block_generic(const u32 *st
 
 	memcpy(&stream[0], &x[0], 16);
 	memcpy(&stream[4], &x[12], 16);
+
+	memzero_explicit(x, sizeof(x));
 }
 EXPORT_SYMBOL(hchacha_block_generic);
author maintainer
#440 lore raw

PATCH 414/491 wifi: rt2x00usb: fix devres lifetime

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johan Hovold <[email protected]>

commit 25369b22223d1c56e42a0cd4ac9137349d5a898e upstream.

USB drivers bind to USB interfaces and any device managed resources should have their lifetime tied to the interface rather than parent USB device. This avoids issues like memory leaks when drivers are unbound without their devices being physically disconnected (e.g. on probe deferral or configuration changes).

Fix the USB anchor lifetime so that it is released on driver unbind.

---
 drivers/net/wireless/ralink/rt2x00/rt2x00usb.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/ralink/rt2x00/rt2x00usb.c +1 -1
@@ -830,7 +830,7 @@ int rt2x00usb_probe(struct usb_interface
 	if (retval)
 		goto exit_free_device;
 
-	rt2x00dev->anchor = devm_kmalloc(&usb_dev->dev,
+	rt2x00dev->anchor = devm_kmalloc(&usb_intf->dev,
 					sizeof(struct usb_anchor),
 					GFP_KERNEL);
 	if (!rt2x00dev->anchor) {
author maintainer
#441 lore raw

PATCH 415/491 xfrm_user: fix info leak in build_report()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Greg Kroah-Hartman <[email protected]>

commit d10119968d0e1f2b669604baf2a8b5fdb72fa6b4 upstream.

struct xfrm_user_report is a __u8 proto field followed by a struct xfrm_selector which means there is three "empty" bytes of padding, but the padding is never zeroed before copying to userspace. Fix that up by zeroing the structure before setting individual member variables.

Cc stable <[email protected]>
Cc Steffen Klassert <[email protected]>
Cc Herbert Xu <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Eric Dumazet <[email protected]>
Cc Jakub Kicinski <[email protected]>
Cc Paolo Abeni <[email protected]>
Cc Simon Horman <[email protected]>
Assisted-by gregkh_clanker_t1000
---
 net/xfrm/xfrm_user.c |    1 +
 1 file changed, 1 insertion(+)
net/xfrm/xfrm_user.c +1 -0
@@ -3445,6 +3445,7 @@ static int build_report(struct sk_buff *
 		return -EMSGSIZE;
 
 	ur = nlmsg_data(nlh);
+	memset(ur, 0, sizeof(*ur));
 	ur->proto = proto;
 	memcpy(&ur->sel, sel, sizeof(ur->sel));
 
author maintainer
#442 lore raw

PATCH 416/491 Input: uinput - fix circular locking dependency with ff-core

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mikhail Gavrilov <[email protected]>

commit 4cda78d6f8bf2b700529f2fbccb994c3e826d7c2 upstream.

A lockdep circular locking dependency warning can be triggered reproducibly when using a force-feedback gamepad with uinput (for example, playing ELDEN RING under Wine with a Flydigi Vader 5 controller):

ff->mutex -> udev->mutex -> input_mutex -> dev->mutex -> ff->mutex

The cycle is caused by four lock acquisition paths:

1. ff upload: input_ff_upload() holds ff->mutex and calls uinput_dev_upload_effect() -> uinput_request_submit() -> uinput_request_send(), which acquires udev->mutex.

2. device create: uinput_ioctl_handler() holds udev->mutex and calls uinput_create_device() -> input_register_device(), which acquires input_mutex.

3. device register: input_register_device() holds input_mutex and calls kbd_connect() -> input_register_handle(), which acquires dev->mutex.

4. evdev release: evdev_release() calls input_flush_device() under dev->mutex, which calls input_ff_flush() acquiring ff->mutex.

Fix this by introducing a new state_lock spinlock to protect udev->state and udev->dev access in uinput_request_send() instead of acquiring udev->mutex. The function only needs to atomically check device state and queue an input event into the ring buffer via uinput_dev_event() -- both operations are safe under a spinlock (ktime_get_ts64() and wake_up_interruptible() do not sleep). This breaks the ff->mutex -> udev->mutex link since a spinlock is a leaf in the lock ordering and cannot form cycles with mutexes.

To keep state transitions visible to uinput_request_send(), protect writes to udev->state in uinput_create_device() and uinput_destroy_device() with the same state_lock spinlock.

Additionally, move init_completion(&request->done) from uinput_request_send() to uinput_request_submit() before uinput_request_reserve_slot(). Once the slot is allocated, uinput_flush_requests() may call complete() on it at any time from the destroy path, so the completion must be initialised before the request becomes visible.

Lock ordering after the fix:

ff->mutex -> state_lock (spinlock, leaf)
udev->mutex -> state_lock (spinlock, leaf)
udev->mutex -> input_mutex -> dev->mutex -> ff->mutex (no back-edge)

---
 drivers/input/misc/uinput.c |   28 +++++++++++++++++++++-------
 1 file changed, 21 insertions(+), 7 deletions(-)
drivers/input/misc/uinput.c +21 -7
@@ -56,6 +56,7 @@ struct uinput_device {
 	struct input_dev	*dev;
 	struct mutex		mutex;
 	enum uinput_state	state;
+	spinlock_t		state_lock;
 	wait_queue_head_t	waitq;
 	unsigned char		ready;
 	unsigned char		head;
@@ -145,19 +146,15 @@ static void uinput_request_release_slot(
 static int uinput_request_send(struct uinput_device *udev,
 			       struct uinput_request *request)
 {
-	int retval;
+	int retval = 0;
 
-	retval = mutex_lock_interruptible(&udev->mutex);
-	if (retval)
-		return retval;
+	spin_lock(&udev->state_lock);
 
 	if (udev->state != UIST_CREATED) {
 		retval = -ENODEV;
 		goto out;
 	}
 
-	init_completion(&request->done);
-
 	/*
 	 * Tell our userspace application about this new request
 	 * by queueing an input event.
@@ -165,7 +162,7 @@ static int uinput_request_send(struct ui
 	uinput_dev_event(udev->dev, EV_UINPUT, request->code, request->id);
 
  out:
-	mutex_unlock(&udev->mutex);
+	spin_unlock(&udev->state_lock);
 	return retval;
 }
 
@@ -174,6 +171,13 @@ static int uinput_request_submit(struct
 {
 	int retval;
 
+	/*
+	 * Initialize completion before allocating the request slot.
+	 * Once the slot is allocated, uinput_flush_requests() may
+	 * complete it at any time, so it must be initialized first.
+	 */
+	init_completion(&request->done);
+
 	retval = uinput_request_reserve_slot(udev, request);
 	if (retval)
 		return retval;
@@ -288,7 +292,14 @@ static void uinput_destroy_device(struct
 	struct input_dev *dev = udev->dev;
 	enum uinput_state old_state = udev->state;
 
+	/*
+	 * Update state under state_lock so that concurrent
+	 * uinput_request_send() sees the state change before we
+	 * flush pending requests and tear down the device.
+	 */
+	spin_lock(&udev->state_lock);
 	udev->state = UIST_NEW_DEVICE;
+	spin_unlock(&udev->state_lock);
 
 	if (dev) {
 		name = dev->name;
@@ -365,7 +376,9 @@ static int uinput_create_device(struct u
 	if (error)
 		goto fail2;
 
+	spin_lock(&udev->state_lock);
 	udev->state = UIST_CREATED;
+	spin_unlock(&udev->state_lock);
 
 	return 0;
 
@@ -383,6 +396,7 @@ static int uinput_open(struct inode *ino
 		return -ENOMEM;
 
 	mutex_init(&newdev->mutex);
+	spin_lock_init(&newdev->state_lock);
 	spin_lock_init(&newdev->requests_lock);
 	init_waitqueue_head(&newdev->requests_waitq);
 	init_waitqueue_head(&newdev->waitq);
author maintainer
#443 lore raw

PATCH 417/491 Input: uinput - take event lock when submitting FF request "event"

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dmitry Torokhov <[email protected]>

commit ff14dafde15c11403fac61367a34fea08926e9ee upstream.

To avoid racing with FF playback events and corrupting device's event queue take event_lock spinlock when calling uinput_dev_event() when submitting a FF upload or erase "event".

---
 drivers/input/misc/uinput.c |    7 +++++++
 1 file changed, 7 insertions(+)
drivers/input/misc/uinput.c +7 -0
@@ -25,8 +25,10 @@
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/fs.h>
+#include <linux/lockdep.h>
 #include <linux/miscdevice.h>
 #include <linux/overflow.h>
+#include <linux/spinlock.h>
 #include <linux/input/mt.h>
 #include "../input-compat.h"
 
@@ -75,6 +77,8 @@ static int uinput_dev_event(struct input
 	struct uinput_device	*udev = input_get_drvdata(dev);
 	struct timespec64	ts;
 
+	lockdep_assert_held(&dev->event_lock);
+
 	ktime_get_ts64(&ts);
 
 	udev->buff[udev->head] = (struct input_event) {
@@ -146,6 +150,7 @@ static void uinput_request_release_slot(
 static int uinput_request_send(struct uinput_device *udev,
 			       struct uinput_request *request)
 {
+	unsigned long flags;
 	int retval = 0;
 
 	spin_lock(&udev->state_lock);
@@ -159,7 +164,9 @@ static int uinput_request_send(struct ui
 	 * Tell our userspace application about this new request
 	 * by queueing an input event.
 	 */
+	spin_lock_irqsave(&udev->dev->event_lock, flags);
 	uinput_dev_event(udev->dev, EV_UINPUT, request->code, request->id);
+	spin_unlock_irqrestore(&udev->dev->event_lock, flags);
 
  out:
 	spin_unlock(&udev->state_lock);
author maintainer
#444 lore raw

PATCH 418/491 mm/hugetlb: fix skipping of unsharing of pmd page tables

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Hildenbrand (Arm) <[email protected]>

In the 5.10 backport of commit b30c14cd6102 ("hugetlb: unshare some PMDs when splitting VMAs") we seemed to have missed that huge_pmd_unshare() still adjusts the address itself.

For this reason, commit 6dfeaff93be1 ("hugetlb/userfaultfd: unshare all pmds for hugetlbfs when register wp") explicitly handled this case by passing a temporary variable instead.

Fix it in 5.10 by doing the same thing.

Fixes f1082f5f3d02 ("hugetlb: unshare some PMDs when splitting VMAs")
---
 mm/hugetlb.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
mm/hugetlb.c +4 -1
index 8efe35ea0baa7..99a71943c1f69 100644
@@ -5787,11 +5787,14 @@ static void hugetlb_unshare_pmds(struct vm_area_struct *vma,
 		i_mmap_assert_write_locked(vma->vm_file->f_mapping);
 	}
 	for (address = start; address < end; address += PUD_SIZE) {
+		unsigned long tmp = address;
+
 		ptep = huge_pte_offset(mm, address, sz);
 		if (!ptep)
 			continue;
 		ptl = huge_pte_lock(h, mm, ptep);
-		huge_pmd_unshare(mm, vma, &address, ptep);
+		/* We don't want 'address' to be changed */
+		huge_pmd_unshare(mm, vma, &tmp, ptep);
 		spin_unlock(ptl);
 	}
 	flush_hugetlb_tlb_range(vma, start, end);
Signature · 1 line
2.53.0
author maintainer
#445 lore raw

PATCH 419/491 mm/hugetlb: make detecting shared pte more reliable

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miaohe Lin <[email protected]>

If the pagetables are shared, we shouldn't copy or take references. Since src could have unshared and dst shares with another vma, huge_pte_none() is thus used to determine whether dst_pte is shared. But this check isn't reliable. A shared pte could have pte none in pagetable in fact. The page count of ptep page should be checked here in order to reliably determine whether pte is shared.

[[email protected]: remove unused local variable dst_entry in copy_hugetlb_page_range()]
Cc Muchun Song <[email protected]>
(cherry picked from commit 3aa4ed8040e1535d95c03cef8b52cf11bf0d8546)

[ David: We don't have 4eae4efa2c29 ("hugetlb: do early cow when page
pinned on src mm", so there are some contextual conflicts. ]

---
 mm/hugetlb.c | 19 +++++++------------
 1 file changed, 7 insertions(+), 12 deletions(-)
mm/hugetlb.c +7 -12
index 99a71943c1f69..a2cab8f2190f8 100644
@@ -3827,7 +3827,7 @@ static bool is_hugetlb_entry_hwpoisoned(pte_t pte)
 int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 			    struct vm_area_struct *vma)
 {
-	pte_t *src_pte, *dst_pte, entry, dst_entry;
+	pte_t *src_pte, *dst_pte, entry;
 	struct page *ptepage;
 	unsigned long addr;
 	int cow;
@@ -3867,27 +3867,22 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 
 		/*
 		 * If the pagetables are shared don't copy or take references.
-		 * dst_pte == src_pte is the common case of src/dest sharing.
 		 *
+		 * dst_pte == src_pte is the common case of src/dest sharing.
 		 * However, src could have 'unshared' and dst shares with
-		 * another vma.  If dst_pte !none, this implies sharing.
-		 * Check here before taking page table lock, and once again
-		 * after taking the lock below.
+		 * another vma. So page_count of ptep page is checked instead
+		 * to reliably determine whether pte is shared.
 		 */
-		dst_entry = huge_ptep_get(dst_pte);
-		if ((dst_pte == src_pte) || !huge_pte_none(dst_entry))
+		if (page_count(virt_to_page(dst_pte)) > 1)
 			continue;
 
 		dst_ptl = huge_pte_lock(h, dst, dst_pte);
 		src_ptl = huge_pte_lockptr(h, src, src_pte);
 		spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING);
 		entry = huge_ptep_get(src_pte);
-		dst_entry = huge_ptep_get(dst_pte);
-		if (huge_pte_none(entry) || !huge_pte_none(dst_entry)) {
+		if (huge_pte_none(entry)) {
 			/*
-			 * Skip if src entry none.  Also, skip in the
-			 * unlikely case dst entry !none as this implies
-			 * sharing with another vma.
+			 * Skip if src entry none.
 			 */
 			;
 		} else if (unlikely(is_hugetlb_entry_migration(entry) ||
Signature · 1 line
2.53.0
#446 lore raw

Re: [PATCH 5.10 419/491] mm/hugetlb: make detecting shared pte more reliable

Quoted 5 lines · Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miaohe Lin <[email protected]>

Missing "commit 3aa4ed8040e1535d95c03cef8b52cf11bf0d8546 upstream."

Ben.

Quoted 71 lines

If the pagetables are shared, we shouldn't copy or take references. Since src could have unshared and dst shares with another vma, huge_pte_none() is thus used to determine whether dst_pte is shared. But this check isn't reliable. A shared pte could have pte none in pagetable in fact. The page count of ptep page should be checked here in order to reliably determine whether pte is shared.

[[email protected]: remove unused local variable dst_entry in copy_hugetlb_page_range()]
Cc Muchun Song <[email protected]>
(cherry picked from commit 3aa4ed8040e1535d95c03cef8b52cf11bf0d8546)

[ David: We don't have 4eae4efa2c29 ("hugetlb: do early cow when page
pinned on src mm", so there are some contextual conflicts. ]

---
 mm/hugetlb.c | 19 +++++++------------
 1 file changed, 7 insertions(+), 12 deletions(-)
mm/hugetlb.c +7 -12
index 99a71943c1f69..a2cab8f2190f8 100644
@@ -3827,7 +3827,7 @@ static bool is_hugetlb_entry_hwpoisoned(pte_t pte)
 int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 			    struct vm_area_struct *vma)
 {
-	pte_t *src_pte, *dst_pte, entry, dst_entry;
+	pte_t *src_pte, *dst_pte, entry;
 	struct page *ptepage;
 	unsigned long addr;
 	int cow;
@@ -3867,27 +3867,22 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 
 		/*
 		 * If the pagetables are shared don't copy or take references.
-		 * dst_pte == src_pte is the common case of src/dest sharing.
 		 *
+		 * dst_pte == src_pte is the common case of src/dest sharing.
 		 * However, src could have 'unshared' and dst shares with
-		 * another vma.  If dst_pte !none, this implies sharing.
-		 * Check here before taking page table lock, and once again
-		 * after taking the lock below.
+		 * another vma. So page_count of ptep page is checked instead
+		 * to reliably determine whether pte is shared.
 		 */
-		dst_entry = huge_ptep_get(dst_pte);
-		if ((dst_pte == src_pte) || !huge_pte_none(dst_entry))
+		if (page_count(virt_to_page(dst_pte)) > 1)
 			continue;
 
 		dst_ptl = huge_pte_lock(h, dst, dst_pte);
 		src_ptl = huge_pte_lockptr(h, src, src_pte);
 		spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING);
 		entry = huge_ptep_get(src_pte);
-		dst_entry = huge_ptep_get(dst_pte);
-		if (huge_pte_none(entry) || !huge_pte_none(dst_entry)) {
+		if (huge_pte_none(entry)) {
 			/*
-			 * Skip if src entry none.  Also, skip in the
-			 * unlikely case dst entry !none as this implies
-			 * sharing with another vma.
+			 * Skip if src entry none.
 			 */
 			;
 		} else if (unlikely(is_hugetlb_entry_migration(entry) ||
Signature · 2 lines
Ben Hutchings
When in doubt, use brute force. - Ken Thompson
maintainer
#447 lore raw

Re: [PATCH 5.10 419/491] mm/hugetlb: make detecting shared pte more reliable

Quoted 1 line · Ben Hutchings

Missing "commit 3aa4ed8040e1535d95c03cef8b52cf11bf0d8546 upstream."

Added the missing "commit XXX upstream." lines to this patch and the other 4 patches in the hugetlb/rmap backport series, thanks.

author maintainer
#448 lore raw

PATCH 420/491 mm/hugetlb: fix copy_hugetlb_page_range() to use ->pt_share_count

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jane Chu <[email protected]>

commit 59d9094df3d79 ("mm: hugetlb: independent PMD page table shared count") introduced ->pt_share_count dedicated to hugetlb PMD share count tracking, but omitted fixing copy_hugetlb_page_range(), leaving the function relying on page_count() for tracking that no longer works.

When lazy page table copy for hugetlb is disabled, that is, revert commit bcd51a3c679d ("hugetlb: lazy page table copies in fork()") fork()'ing with hugetlb PMD sharing quickly lockup -

[  239.446559] watchdog: BUG: soft lockup - CPU#75 stuck for 27s!
[  239.446611] RIP: 0010:native_queued_spin_lock_slowpath+0x7e/0x2e0
[  239.446631] Call Trace:
[  239.446633]  <TASK>
[  239.446636]  _raw_spin_lock+0x3f/0x60
[  239.446639]  copy_hugetlb_page_range+0x258/0xb50
[  239.446645]  copy_page_range+0x22b/0x2c0
[  239.446651]  dup_mmap+0x3e2/0x770
[  239.446654]  dup_mm.constprop.0+0x5e/0x230
[  239.446657]  copy_process+0xd17/0x1760
[  239.446660]  kernel_clone+0xc0/0x3e0
[  239.446661]  __do_sys_clone+0x65/0xa0
[  239.446664]  do_syscall_64+0x82/0x930
[  239.446668]  ? count_memcg_events+0xd2/0x190
[  239.446671]  ? syscall_trace_enter+0x14e/0x1f0
[  239.446676]  ? syscall_exit_work+0x118/0x150
[  239.446677]  ? arch_exit_to_user_mode_prepare.constprop.0+0x9/0xb0
[  239.446681]  ? clear_bhb_loop+0x30/0x80
[  239.446684]  ? clear_bhb_loop+0x30/0x80
[  239.446686]  entry_SYSCALL_64_after_hwframe+0x76/0x7e

There are two options to resolve the potential latent issue:
1. warn against PMD sharing in copy_hugetlb_page_range(),
2. fix it.
This patch opts for the second option.
While at it, simplify the comment, the details are not actually relevant anymore.

Fixes 59d9094df3d7 ("mm: hugetlb: independent PMD page table shared count")
Cc Jann Horn <[email protected]>
Cc Liu Shixin <[email protected]>
Cc Muchun Song <[email protected]>
(cherry picked from commit 14967a9c7d247841b0312c48dcf8cd29e55a4cc8)

[ David: We don't have ptdesc and the wrappers, so work directly on the
page->pt_share_count. CONFIG_HUGETLB_PMD_PAGE_TABLE_SHARING is still called CONFIG_ARCH_WANT_HUGE_PMD_SHARE. ]

---
 mm/hugetlb.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)
mm/hugetlb.c +4 -9
index a2cab8f2190f8..8fa34032bc173 100644
@@ -3865,16 +3865,11 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 			break;
 		}
 
-		/*
-		 * If the pagetables are shared don't copy or take references.
-		 *
-		 * dst_pte == src_pte is the common case of src/dest sharing.
-		 * However, src could have 'unshared' and dst shares with
-		 * another vma. So page_count of ptep page is checked instead
-		 * to reliably determine whether pte is shared.
-		 */
-		if (page_count(virt_to_page(dst_pte)) > 1)
+#ifdef CONFIG_ARCH_WANT_HUGE_PMD_SHARE
+		/* If the pagetables are shared, there is nothing to do */
+		if (atomic_read(&virt_to_page(dst_pte)->pt_share_count))
 			continue;
+#endif
 
 		dst_ptl = huge_pte_lock(h, dst, dst_pte);
 		src_ptl = huge_pte_lockptr(h, src, src_pte);
Signature · 1 line
2.53.0
#449 lore raw

Re: [PATCH 5.10 420/491] mm/hugetlb: fix copy_hugetlb_page_range() to use ->pt_share_count

Quoted 5 lines · Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jane Chu <[email protected]>

This one was commit 14967a9c7d247841b0312c48dcf8cd29e55a4cc8 upstream.

Ben.

Quoted 82 lines

commit 59d9094df3d79 ("mm: hugetlb: independent PMD page table shared count") introduced ->pt_share_count dedicated to hugetlb PMD share count tracking, but omitted fixing copy_hugetlb_page_range(), leaving the function relying on page_count() for tracking that no longer works.

When lazy page table copy for hugetlb is disabled, that is, revert commit bcd51a3c679d ("hugetlb: lazy page table copies in fork()") fork()'ing with hugetlb PMD sharing quickly lockup -

[  239.446559] watchdog: BUG: soft lockup - CPU#75 stuck for 27s!
[  239.446611] RIP: 0010:native_queued_spin_lock_slowpath+0x7e/0x2e0
[  239.446631] Call Trace:
[  239.446633]  <TASK>
[  239.446636]  _raw_spin_lock+0x3f/0x60
[  239.446639]  copy_hugetlb_page_range+0x258/0xb50
[  239.446645]  copy_page_range+0x22b/0x2c0
[  239.446651]  dup_mmap+0x3e2/0x770
[  239.446654]  dup_mm.constprop.0+0x5e/0x230
[  239.446657]  copy_process+0xd17/0x1760
[  239.446660]  kernel_clone+0xc0/0x3e0
[  239.446661]  __do_sys_clone+0x65/0xa0
[  239.446664]  do_syscall_64+0x82/0x930
[  239.446668]  ? count_memcg_events+0xd2/0x190
[  239.446671]  ? syscall_trace_enter+0x14e/0x1f0
[  239.446676]  ? syscall_exit_work+0x118/0x150
[  239.446677]  ? arch_exit_to_user_mode_prepare.constprop.0+0x9/0xb0
[  239.446681]  ? clear_bhb_loop+0x30/0x80
[  239.446684]  ? clear_bhb_loop+0x30/0x80
[  239.446686]  entry_SYSCALL_64_after_hwframe+0x76/0x7e

There are two options to resolve the potential latent issue:
1. warn against PMD sharing in copy_hugetlb_page_range(),
2. fix it.
This patch opts for the second option.
While at it, simplify the comment, the details are not actually relevant anymore.

Fixes 59d9094df3d7 ("mm: hugetlb: independent PMD page table shared count")
Cc Jann Horn <[email protected]>
Cc Liu Shixin <[email protected]>
Cc Muchun Song <[email protected]>
(cherry picked from commit 14967a9c7d247841b0312c48dcf8cd29e55a4cc8)

[ David: We don't have ptdesc and the wrappers, so work directly on the
page->pt_share_count. CONFIG_HUGETLB_PMD_PAGE_TABLE_SHARING is still called CONFIG_ARCH_WANT_HUGE_PMD_SHARE. ]

---
 mm/hugetlb.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)
mm/hugetlb.c +4 -9
index a2cab8f2190f8..8fa34032bc173 100644
@@ -3865,16 +3865,11 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 			break;
 		}
 
-		/*
-		 * If the pagetables are shared don't copy or take references.
-		 *
-		 * dst_pte == src_pte is the common case of src/dest sharing.
-		 * However, src could have 'unshared' and dst shares with
-		 * another vma. So page_count of ptep page is checked instead
-		 * to reliably determine whether pte is shared.
-		 */
-		if (page_count(virt_to_page(dst_pte)) > 1)
+#ifdef CONFIG_ARCH_WANT_HUGE_PMD_SHARE
+		/* If the pagetables are shared, there is nothing to do */
+		if (atomic_read(&virt_to_page(dst_pte)->pt_share_count))
 			continue;
+#endif
 
 		dst_ptl = huge_pte_lock(h, dst, dst_pte);
 		src_ptl = huge_pte_lockptr(h, src, src_pte);
Signature · 2 lines
Ben Hutchings
When in doubt, use brute force. - Ken Thompson
author maintainer
#450 lore raw

PATCH 421/491 mm/hugetlb: fix hugetlb_pmd_shared()

In reply to Greg Kroah-Hartman

5.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Hildenbrand (Red Hat) <[email protected]>

Commit ca1a47cd3f5f4c46ca188b1c9a27af87d1ab2216 upstream.

Patch series "mm/hugetlb: fixes for PMD table sharing (incl. using mmu_gather)", v3.

One functional fix, one performance regression fix, and two related comment fixes.

I cleaned up my prototype I recently shared [1] for the performance fix, deferring most of the cleanups I had in the prototype to a later point. While doing that I identified the other things.

The goal of this patch set is to be backported to stable trees "fairly" easily. At least patch #1 and #4.

Patch #1 fixes hugetlb_pmd_shared() not detecting any sharing Patch #2 + #3 are simple comment fixes that patch #4 interacts with. Patch #4 is a fix for the reported performance regression due to excessive IPI broadcasts during fork()+exit().

The last patch is all about TLB flushes, IPIs and mmu_gather.
Read: complicated

There are plenty of cleanups in the future to be had + one reasonable optimization on x86. But that's all out of scope for this series.

Runtime tested, with a focus on fixing the performance regression using the original reproducer [2] on x86.

This patch (of 4):

We switched from (wrongly) using the page count to an independent shared count. Now, shared page tables have a refcount of 1 (excluding speculative references) and instead use ptdesc->pt_share_count to identify sharing.

We didn't convert hugetlb_pmd_shared(), so right now, we would never detect a shared PMD table as such, because sharing/unsharing no longer touches the refcount of a PMD table.

Page migration, like mbind() or migrate_pages() would allow for migrating folios mapped into such shared PMD tables, even though the folios are not exclusive. In smaps we would account them as "private" although they are "shared", and we would be wrongly setting the PM_MMAP_EXCLUSIVE in the pagemap interface.

Fix it by properly using ptdesc_pmd_is_shared() in hugetlb_pmd_shared().

[ David: We don't have ptdesc and the wrappers, so work directly on
page->pt_share_count. ]

---
 include/linux/hugetlb.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/hugetlb.h +1 -1
index 1c03935aa3d13..0d3fece27031c 100644
@@ -955,7 +955,7 @@ static inline __init void hugetlb_cma_check(void)
 #ifdef CONFIG_ARCH_WANT_HUGE_PMD_SHARE
 static inline bool hugetlb_pmd_shared(pte_t *pte)
 {
-	return page_count(virt_to_page(pte)) > 1;
+	return atomic_read(&virt_to_page(pte)->pt_share_count);
 }
 #else
 static inline bool hugetlb_pmd_shared(pte_t *pte)
Signature · 1 line
2.53.0