March 2025 - Linux-stable-mirror

[PATCH AUTOSEL 6.1 1/9] HID: pidff: Convert infinite length from Linux API to PID standard

by Sasha Levin

From: Tomasz Pakuła <tomasz.pakula.oficjalny(a)gmail.com> [ Upstream commit 37e0591fe44dce39d1ebc7a82d5b6e4dba1582eb ] Software uses 0 as de-facto infinite lenght on Linux FF apis (SDL), Linux doesn't actually define anythi as of now, while USB PID defines NULL (0xffff). Most PID devices do not expect a 0-length effect and can't interpret it as infinite. This change fixes Force Feedback for most PID compliant devices. As most games depend on updating the values of already playing infinite effects, this is crucial to ensure they will actually work. Previously, users had to rely on third-party software to do this conversion and make their PID devices usable. Co-developed-by: Makarenko Oleg <oleg(a)makarenk.ooo> Signed-off-by: Makarenko Oleg <oleg(a)makarenk.ooo> Signed-off-by: Tomasz Pakuła <tomasz.pakula.oficjalny(a)gmail.com> Reviewed-by: Michał Kopeć <michal(a)nozomi.space> Reviewed-by: Paul Dino Jones <paul(a)spacefreak18.xyz> Tested-by: Paul Dino Jones <paul(a)spacefreak18.xyz> Tested-by: Cristóferson Bueno <cbueno81(a)gmail.com> Tested-by: Pablo Cisneros <patchkez(a)protonmail.com> Signed-off-by: Jiri Kosina <jkosina(a)suse.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- drivers/hid/usbhid/hid-pidff.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/hid/usbhid/hid-pidff.c b/drivers/hid/usbhid/hid-pidff.c index 3b4ee21cd8111..5fe4422bb5bad 100644 --- a/drivers/hid/usbhid/hid-pidff.c +++ b/drivers/hid/usbhid/hid-pidff.c @@ -21,6 +21,7 @@ #include "usbhid.h" #define PID_EFFECTS_MAX 64 +#define PID_INFINITE 0xffff /* Report usage table used to put reports into an array */ @@ -301,7 +302,12 @@ static void pidff_set_effect_report(struct pidff_device *pidff, pidff->block_load[PID_EFFECT_BLOCK_INDEX].value[0]; pidff->set_effect_type->value[0] = pidff->create_new_effect_type->value[0]; - pidff->set_effect[PID_DURATION].value[0] = effect->replay.length; + + /* Convert infinite length from Linux API (0) + to PID standard (NULL) if needed */ + pidff->set_effect[PID_DURATION].value[0] = + effect->replay.length == 0 ? PID_INFINITE : effect->replay.length; + pidff->set_effect[PID_TRIGGER_BUTTON].value[0] = effect->trigger.button; pidff->set_effect[PID_TRIGGER_REPEAT_INT].value[0] = effect->trigger.interval; -- 2.39.5

1 month, 3 weeks

2
9
0 0

[PATCH AUTOSEL 5.10 1/5] HID: pidff: Convert infinite length from Linux API to PID standard

by Sasha Levin

From: Tomasz Pakuła <tomasz.pakula.oficjalny(a)gmail.com> [ Upstream commit 37e0591fe44dce39d1ebc7a82d5b6e4dba1582eb ] Software uses 0 as de-facto infinite lenght on Linux FF apis (SDL), Linux doesn't actually define anythi as of now, while USB PID defines NULL (0xffff). Most PID devices do not expect a 0-length effect and can't interpret it as infinite. This change fixes Force Feedback for most PID compliant devices. As most games depend on updating the values of already playing infinite effects, this is crucial to ensure they will actually work. Previously, users had to rely on third-party software to do this conversion and make their PID devices usable. Co-developed-by: Makarenko Oleg <oleg(a)makarenk.ooo> Signed-off-by: Makarenko Oleg <oleg(a)makarenk.ooo> Signed-off-by: Tomasz Pakuła <tomasz.pakula.oficjalny(a)gmail.com> Reviewed-by: Michał Kopeć <michal(a)nozomi.space> Reviewed-by: Paul Dino Jones <paul(a)spacefreak18.xyz> Tested-by: Paul Dino Jones <paul(a)spacefreak18.xyz> Tested-by: Cristóferson Bueno <cbueno81(a)gmail.com> Tested-by: Pablo Cisneros <patchkez(a)protonmail.com> Signed-off-by: Jiri Kosina <jkosina(a)suse.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- drivers/hid/usbhid/hid-pidff.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/hid/usbhid/hid-pidff.c b/drivers/hid/usbhid/hid-pidff.c index 07a9fe97d2e05..badcb5f28607e 100644 --- a/drivers/hid/usbhid/hid-pidff.c +++ b/drivers/hid/usbhid/hid-pidff.c @@ -21,6 +21,7 @@ #include "usbhid.h" #define PID_EFFECTS_MAX 64 +#define PID_INFINITE 0xffff /* Report usage table used to put reports into an array */ @@ -301,7 +302,12 @@ static void pidff_set_effect_report(struct pidff_device *pidff, pidff->block_load[PID_EFFECT_BLOCK_INDEX].value[0]; pidff->set_effect_type->value[0] = pidff->create_new_effect_type->value[0]; - pidff->set_effect[PID_DURATION].value[0] = effect->replay.length; + + /* Convert infinite length from Linux API (0) + to PID standard (NULL) if needed */ + pidff->set_effect[PID_DURATION].value[0] = + effect->replay.length == 0 ? PID_INFINITE : effect->replay.length; + pidff->set_effect[PID_TRIGGER_BUTTON].value[0] = effect->trigger.button; pidff->set_effect[PID_TRIGGER_REPEAT_INT].value[0] = effect->trigger.interval; -- 2.39.5

1 month, 3 weeks

2
5
0 0

[PATCH] iommu/amd: Fix potential buffer overflow in parse_ivrs_acpihid

by Pavel Paklov

There is a string parsing logic error which can lead to an overflow of hid or uid buffers. Comparing ACPIID_LEN against a total string length doesn't take into account the lengths of individual hid and uid buffers so the check is insufficient in some cases. For example if the length of hid string is 4 and the length of the uid string is 260, the length of str will be equal to ACPIID_LEN + 1 but uid string will overflow uid buffer which size is 256. The same applies to the hid string with length 13 and uid string with length 250. Check the length of hid and uid strings separately to prevent buffer overflow. Found by Linux Verification Center (linuxtesting.org) with SVACE. Fixes: ca3bf5d47cec ("iommu/amd: Introduces ivrs_acpihid kernel parameter") Cc: stable(a)vger.kernel.org Signed-off-by: Pavel Paklov <Pavel.Paklov(a)cyberprotect.ru> --- drivers/iommu/amd/init.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/iommu/amd/init.c b/drivers/iommu/amd/init.c index cb536d372b12..fb82f8035c0f 100644 --- a/drivers/iommu/amd/init.c +++ b/drivers/iommu/amd/init.c @@ -3677,6 +3677,14 @@ static int __init parse_ivrs_acpihid(char *str) while (*uid == '0' && *(uid + 1)) uid++; + if (strlen(hid) >= ACPIHID_HID_LEN) { + pr_err("Invalid command line: hid is too long\n"); + return 1; + } else if (strlen(uid) >= ACPIHID_UID_LEN) { + pr_err("Invalid command line: uid is too long\n"); + return 1; + } + i = early_acpihid_map_size++; memcpy(early_acpihid_map[i].hid, hid, strlen(hid)); memcpy(early_acpihid_map[i].uid, uid, strlen(uid)); -- 2.43.0

1 month, 3 weeks

2
1
0 0

FAILED: patch "[PATCH] Fix mmu notifiers for range-based invalidates" failed to apply to 6.6-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.6-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.6.y git checkout FETCH_HEAD git cherry-pick -x f7edb07ad7c66eab3dce57384f33b9799d579133 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025031650-yarn-arrogant-2584@gregkh' --subject-prefix 'PATCH 6.6.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From f7edb07ad7c66eab3dce57384f33b9799d579133 Mon Sep 17 00:00:00 2001 From: Piotr Jaroszynski <pjaroszynski(a)nvidia.com> Date: Tue, 4 Mar 2025 00:51:27 -0800 Subject: [PATCH] Fix mmu notifiers for range-based invalidates Update the __flush_tlb_range_op macro not to modify its parameters as these are unexepcted semantics. In practice, this fixes the call to mmu_notifier_arch_invalidate_secondary_tlbs() in __flush_tlb_range_nosync() to use the correct range instead of an empty range with start=end. The empty range was (un)lucky as it results in taking the invalidate-all path that doesn't cause correctness issues, but can certainly result in suboptimal perf. This has been broken since commit 6bbd42e2df8f ("mmu_notifiers: call invalidate_range() when invalidating TLBs") when the call to the notifiers was added to __flush_tlb_range(). It predates the addition of the __flush_tlb_range_op() macro from commit 360839027a6e ("arm64: tlb: Refactor the core flush algorithm of __flush_tlb_range") that made the bug hard to spot. Fixes: 6bbd42e2df8f ("mmu_notifiers: call invalidate_range() when invalidating TLBs") Signed-off-by: Piotr Jaroszynski <pjaroszynski(a)nvidia.com> Cc: Catalin Marinas <catalin.marinas(a)arm.com> Cc: Will Deacon <will(a)kernel.org> Cc: Robin Murphy <robin.murphy(a)arm.com> Cc: Alistair Popple <apopple(a)nvidia.com> Cc: Raghavendra Rao Ananta <rananta(a)google.com> Cc: SeongJae Park <sj(a)kernel.org> Cc: Jason Gunthorpe <jgg(a)nvidia.com> Cc: John Hubbard <jhubbard(a)nvidia.com> Cc: Nicolin Chen <nicolinc(a)nvidia.com> Cc: linux-arm-kernel(a)lists.infradead.org Cc: iommu(a)lists.linux.dev Cc: linux-mm(a)kvack.org Cc: linux-kernel(a)vger.kernel.org Cc: stable(a)vger.kernel.org Reviewed-by: Catalin Marinas <catalin.marinas(a)arm.com> Reviewed-by: Alistair Popple <apopple(a)nvidia.com> Link: https://lore.kernel.org/r/20250304085127.2238030-1-pjaroszynski@nvidia.com Signed-off-by: Will Deacon <will(a)kernel.org> diff --git a/arch/arm64/include/asm/tlbflush.h b/arch/arm64/include/asm/tlbflush.h index bc94e036a26b..8104aee4f9a0 100644 --- a/arch/arm64/include/asm/tlbflush.h +++ b/arch/arm64/include/asm/tlbflush.h @@ -396,33 +396,35 @@ static inline void arch_tlbbatch_flush(struct arch_tlbflush_unmap_batch *batch) #define __flush_tlb_range_op(op, start, pages, stride, \ asid, tlb_level, tlbi_user, lpa2) \ do { \ + typeof(start) __flush_start = start; \ + typeof(pages) __flush_pages = pages; \ int num = 0; \ int scale = 3; \ int shift = lpa2 ? 16 : PAGE_SHIFT; \ unsigned long addr; \ \ - while (pages > 0) { \ + while (__flush_pages > 0) { \ if (!system_supports_tlb_range() || \ - pages == 1 || \ - (lpa2 && start != ALIGN(start, SZ_64K))) { \ - addr = __TLBI_VADDR(start, asid); \ + __flush_pages == 1 || \ + (lpa2 && __flush_start != ALIGN(__flush_start, SZ_64K))) { \ + addr = __TLBI_VADDR(__flush_start, asid); \ __tlbi_level(op, addr, tlb_level); \ if (tlbi_user) \ __tlbi_user_level(op, addr, tlb_level); \ - start += stride; \ - pages -= stride >> PAGE_SHIFT; \ + __flush_start += stride; \ + __flush_pages -= stride >> PAGE_SHIFT; \ continue; \ } \ \ - num = __TLBI_RANGE_NUM(pages, scale); \ + num = __TLBI_RANGE_NUM(__flush_pages, scale); \ if (num >= 0) { \ - addr = __TLBI_VADDR_RANGE(start >> shift, asid, \ + addr = __TLBI_VADDR_RANGE(__flush_start >> shift, asid, \ scale, num, tlb_level); \ __tlbi(r##op, addr); \ if (tlbi_user) \ __tlbi_user(r##op, addr); \ - start += __TLBI_RANGE_PAGES(num, scale) << PAGE_SHIFT; \ - pages -= __TLBI_RANGE_PAGES(num, scale); \ + __flush_start += __TLBI_RANGE_PAGES(num, scale) << PAGE_SHIFT; \ + __flush_pages -= __TLBI_RANGE_PAGES(num, scale);\ } \ scale--; \ } \

1 month, 3 weeks

2
1
0 0

[PATCH] hfs/hfsplus: fix slab-out-of-bounds in hfs_bnode_read_key

by Vasiliy Kovalev

Syzbot reported an issue in hfs subsystem: BUG: KASAN: slab-out-of-bounds in memcpy_from_page include/linux/highmem.h:423 [inline] BUG: KASAN: slab-out-of-bounds in hfs_bnode_read fs/hfs/bnode.c:35 [inline] BUG: KASAN: slab-out-of-bounds in hfs_bnode_read_key+0x314/0x450 fs/hfs/bnode.c:70 Write of size 94 at addr ffff8880123cd100 by task syz-executor237/5102 Call Trace: <TASK> __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:377 [inline] print_report+0x169/0x550 mm/kasan/report.c:488 kasan_report+0x143/0x180 mm/kasan/report.c:601 kasan_check_range+0x282/0x290 mm/kasan/generic.c:189 __asan_memcpy+0x40/0x70 mm/kasan/shadow.c:106 memcpy_from_page include/linux/highmem.h:423 [inline] hfs_bnode_read fs/hfs/bnode.c:35 [inline] hfs_bnode_read_key+0x314/0x450 fs/hfs/bnode.c:70 hfs_brec_insert+0x7f3/0xbd0 fs/hfs/brec.c:159 hfs_cat_create+0x41d/0xa50 fs/hfs/catalog.c:118 hfs_mkdir+0x6c/0xe0 fs/hfs/dir.c:232 vfs_mkdir+0x2f9/0x4f0 fs/namei.c:4257 do_mkdirat+0x264/0x3a0 fs/namei.c:4280 __do_sys_mkdir fs/namei.c:4300 [inline] __se_sys_mkdir fs/namei.c:4298 [inline] __x64_sys_mkdir+0x6c/0x80 fs/namei.c:4298 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7fbdd6057a99 Add a check for key length in hfs_bnode_read_key to prevent out-of-bounds memory access. If the key length is invalid, the key buffer is cleared, improving stability and reliability. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: syzbot+5f3a973ed3dfb85a6683(a)syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=5f3a973ed3dfb85a6683 Cc: stable(a)vger.kernel.org Signed-off-by: Vasiliy Kovalev <kovalev(a)altlinux.org> --- fs/hfs/bnode.c | 6 ++++++ fs/hfsplus/bnode.c | 6 ++++++ 2 files changed, 12 insertions(+) diff --git a/fs/hfs/bnode.c b/fs/hfs/bnode.c index 6add6ebfef8967..cb823a8a6ba960 100644 --- a/fs/hfs/bnode.c +++ b/fs/hfs/bnode.c @@ -67,6 +67,12 @@ void hfs_bnode_read_key(struct hfs_bnode *node, void *key, int off) else key_len = tree->max_key_len + 1; + if (key_len > sizeof(hfs_btree_key) || key_len < 1) { + memset(key, 0, sizeof(hfs_btree_key)); + pr_err("hfs: Invalid key length: %d\n", key_len); + return; + } + hfs_bnode_read(node, key, off, key_len); } diff --git a/fs/hfsplus/bnode.c b/fs/hfsplus/bnode.c index 87974d5e679156..079ea80534f7de 100644 --- a/fs/hfsplus/bnode.c +++ b/fs/hfsplus/bnode.c @@ -67,6 +67,12 @@ void hfs_bnode_read_key(struct hfs_bnode *node, void *key, int off) else key_len = tree->max_key_len + 2; + if (key_len > sizeof(hfsplus_btree_key) || key_len < 1) { + memset(key, 0, sizeof(hfsplus_btree_key)); + pr_err("hfsplus: Invalid key length: %d\n", key_len); + return; + } + hfs_bnode_read(node, key, off, key_len); } -- 2.33.8

1 month, 4 weeks

10
19
0 0

[PATCH v3] usb: dwc3: Abort suspend on soft disconnect failure

by Kuen-Han Tsai

When dwc3_gadget_soft_disconnect() fails, dwc3_suspend_common() keeps going with the suspend, resulting in a period where the power domain is off, but the gadget driver remains connected. Within this time frame, invoking vbus_event_work() will cause an error as it attempts to access DWC3 registers for endpoint disabling after the power domain has been completely shut down. Abort the suspend sequence when dwc3_gadget_suspend() cannot halt the controller and proceeds with a soft connect. Fixes: 9f8a67b65a49 ("usb: dwc3: gadget: fix gadget suspend/resume") CC: stable(a)vger.kernel.org Signed-off-by: Kuen-Han Tsai <khtsai(a)google.com> --- Kernel panic - not syncing: Asynchronous SError Interrupt Workqueue: events vbus_event_work Call trace: dump_backtrace+0xf4/0x118 show_stack+0x18/0x24 dump_stack_lvl+0x60/0x7c dump_stack+0x18/0x3c panic+0x16c/0x390 nmi_panic+0xa4/0xa8 arm64_serror_panic+0x6c/0x94 do_serror+0xc4/0xd0 el1h_64_error_handler+0x34/0x48 el1h_64_error+0x68/0x6c readl+0x4c/0x8c __dwc3_gadget_ep_disable+0x48/0x230 dwc3_gadget_ep_disable+0x50/0xc0 usb_ep_disable+0x44/0xe4 ffs_func_eps_disable+0x64/0xc8 ffs_func_set_alt+0x74/0x368 ffs_func_disable+0x18/0x28 composite_disconnect+0x90/0xec configfs_composite_disconnect+0x64/0x88 usb_gadget_disconnect_locked+0xc0/0x168 vbus_event_work+0x3c/0x58 process_one_work+0x1e4/0x43c worker_thread+0x25c/0x430 kthread+0x104/0x1d4 ret_from_fork+0x10/0x20 --- Changelog: v3: - change the Fixes tag v2: - move declarations in separate lines - add the Fixes tag --- drivers/usb/dwc3/core.c | 9 +++++++-- drivers/usb/dwc3/gadget.c | 22 +++++++++------------- 2 files changed, 16 insertions(+), 15 deletions(-) diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c index 66a08b527165..1cf1996ae1fb 100644 --- a/drivers/usb/dwc3/core.c +++ b/drivers/usb/dwc3/core.c @@ -2388,6 +2388,7 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg) { u32 reg; int i; + int ret; if (!pm_runtime_suspended(dwc->dev) && !PMSG_IS_AUTO(msg)) { dwc->susphy_state = (dwc3_readl(dwc->regs, DWC3_GUSB2PHYCFG(0)) & @@ -2406,7 +2407,9 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg) case DWC3_GCTL_PRTCAP_DEVICE: if (pm_runtime_suspended(dwc->dev)) break; - dwc3_gadget_suspend(dwc); + ret = dwc3_gadget_suspend(dwc); + if (ret) + return ret synchronize_irq(dwc->irq_gadget); dwc3_core_exit(dwc); break; @@ -2441,7 +2444,9 @@ static int dwc3_suspend_common(struct dwc3 *dwc, pm_message_t msg) break; if (dwc->current_otg_role == DWC3_OTG_ROLE_DEVICE) { - dwc3_gadget_suspend(dwc); + ret = dwc3_gadget_suspend(dwc); + if (ret) + return ret; synchronize_irq(dwc->irq_gadget); } diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index 89a4dc8ebf94..316c1589618e 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -4776,26 +4776,22 @@ int dwc3_gadget_suspend(struct dwc3 *dwc) int ret; ret = dwc3_gadget_soft_disconnect(dwc); - if (ret) - goto err; - - spin_lock_irqsave(&dwc->lock, flags); - if (dwc->gadget_driver) - dwc3_disconnect_gadget(dwc); - spin_unlock_irqrestore(&dwc->lock, flags); - - return 0; - -err: /* * Attempt to reset the controller's state. Likely no * communication can be established until the host * performs a port reset. */ - if (dwc->softconnect) + if (ret && dwc->softconnect) { dwc3_gadget_soft_connect(dwc); + return ret; + } - return ret; + spin_lock_irqsave(&dwc->lock, flags); + if (dwc->gadget_driver) + dwc3_disconnect_gadget(dwc); + spin_unlock_irqrestore(&dwc->lock, flags); + + return 0; } int dwc3_gadget_resume(struct dwc3 *dwc) -- 2.49.0.472.ge94155a9ec-goog

1 month, 4 weeks

3
4
0 0

[PATCH] sched/fair: Add null pointer check to pick_next_entity()

by Pat Cody

pick_eevdf() can return null, resulting in a null pointer dereference crash in pick_next_entity() The other call site of pick_eevdf() can already handle a null pointer, and pick_next_entity() can already return null as well. Add an extra check to handle the null return here. Cc: stable(a)vger.kernel.org Fixes: f12e148892ed ("sched/fair: Prepare pick_next_task() for delayed dequeue") Signed-off-by: Pat Cody <pat(a)patcody.io> --- kernel/sched/fair.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index a553181dc764..f2157298cbce 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5560,6 +5560,8 @@ pick_next_entity(struct rq *rq, struct cfs_rq *cfs_rq) } struct sched_entity *se = pick_eevdf(cfs_rq); + if (!se) + return NULL; if (se->sched_delayed) { dequeue_entities(rq, se, DEQUEUE_SLEEP | DEQUEUE_DELAYED); /* -- 2.47.1

1 month, 4 weeks

5
15
0 0

[PATCH 1/2] PCI/MSI: Add MSIX option to write to ENTRY_DATA before any reads

by dullfire＠yahoo.com

From: Jonathan Currier <dullfire(a)yahoo.com> Commit 7d5ec3d36123 ("PCI/MSI: Mask all unused MSI-X entries") introduces a readl() from ENTRY_VECTOR_CTRL before the writel() to ENTRY_DATA. This is correct, however some hardware, like the Sun Neptune chips, the niu module, will cause an error and/or fatal trap if any MSIX table entry is read before the corresponding ENTRY_DATA field is written to. This patch adds an optional early writel() in msix_prepare_msi_desc(). Cc: stable(a)vger.kernel.org Signed-off-by: Jonathan Currier <dullfire(a)yahoo.com> --- drivers/pci/msi/msi.c | 2 ++ include/linux/pci.h | 2 ++ 2 files changed, 4 insertions(+) diff --git a/drivers/pci/msi/msi.c b/drivers/pci/msi/msi.c index 3a45879d85db..50d87fb5e37f 100644 --- a/drivers/pci/msi/msi.c +++ b/drivers/pci/msi/msi.c @@ -611,6 +611,8 @@ void msix_prepare_msi_desc(struct pci_dev *dev, struct msi_desc *desc) if (desc->pci.msi_attrib.can_mask) { void __iomem *addr = pci_msix_desc_addr(desc); + if (dev->dev_flags & PCI_DEV_FLAGS_MSIX_TOUCH_ENTRY_DATA_FIRST) + writel(0, addr + PCI_MSIX_ENTRY_DATA); desc->pci.msix_ctrl = readl(addr + PCI_MSIX_ENTRY_VECTOR_CTRL); } } diff --git a/include/linux/pci.h b/include/linux/pci.h index 37d97bef060f..b8b95b58d522 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -245,6 +245,8 @@ enum pci_dev_flags { PCI_DEV_FLAGS_NO_RELAXED_ORDERING = (__force pci_dev_flags_t) (1 << 11), /* Device does honor MSI masking despite saying otherwise */ PCI_DEV_FLAGS_HAS_MSI_MASKING = (__force pci_dev_flags_t) (1 << 12), + /* Device requires write to PCI_MSIX_ENTRY_DATA before any MSIX reads */ + PCI_DEV_FLAGS_MSIX_TOUCH_ENTRY_DATA_FIRST = (__force pci_dev_flags_t) (1 << 13), }; enum pci_irq_reroute_variant { -- 2.45.2

1 month, 4 weeks

5
7
0 0

[PATCH 2/2] net/niu: niu requires MSIX ENTRY_DATA fields touch before entry reads

by dullfire＠yahoo.com

From: Jonathan Currier <dullfire(a)yahoo.com> Fix niu_try_msix() to not cause a fatal trap on sparc systems. Set PCI_DEV_FLAGS_MSIX_TOUCH_ENTRY_DATA_FIRST on the struct pci_dev to work around a bug in the hardware or firmware. For each vector entry in the msix table, niu chips will cause a fatal trap if any registers in that entry are read before that entries' ENTRY_DATA register is written to. Testing indicates writes to other registers are not sufficient to prevent the fatal trap, however the value does not appear to matter. This only needs to happen once after power up, so simply rebooting into a kernel lacking this fix will NOT cause the trap. NON-RESUMABLE ERROR: Reporting on cpu 64 NON-RESUMABLE ERROR: TPC [0x00000000005f6900] <msix_prepare_msi_desc+0x90/0xa0> NON-RESUMABLE ERROR: RAW [4010000000000016:00000e37f93e32ff:0000000202000080:ffffffffffffffff NON-RESUMABLE ERROR: 0000000800000000:0000000000000000:0000000000000000:0000000000000000] NON-RESUMABLE ERROR: handle [0x4010000000000016] stick [0x00000e37f93e32ff] NON-RESUMABLE ERROR: type [precise nonresumable] NON-RESUMABLE ERROR: attrs [0x02000080] < ASI sp-faulted priv > NON-RESUMABLE ERROR: raddr [0xffffffffffffffff] NON-RESUMABLE ERROR: insn effective address [0x000000c50020000c] NON-RESUMABLE ERROR: size [0x8] NON-RESUMABLE ERROR: asi [0x00] CPU: 64 UID: 0 PID: 745 Comm: kworker/64:1 Not tainted 6.11.5 #63 Workqueue: events work_for_cpu_fn TSTATE: 0000000011001602 TPC: 00000000005f6900 TNPC: 00000000005f6904 Y: 00000000 Not tainted TPC: <msix_prepare_msi_desc+0x90/0xa0> g0: 00000000000002e9 g1: 000000000000000c g2: 000000c50020000c g3: 0000000000000100 g4: ffff8000470307c0 g5: ffff800fec5be000 g6: ffff800047a08000 g7: 0000000000000000 o0: ffff800014feb000 o1: ffff800047a0b620 o2: 0000000000000011 o3: ffff800047a0b620 o4: 0000000000000080 o5: 0000000000000011 sp: ffff800047a0ad51 ret_pc: 00000000005f7128 RPC: <__pci_enable_msix_range+0x3cc/0x460> l0: 000000000000000d l1: 000000000000c01f l2: ffff800014feb0a8 l3: 0000000000000020 l4: 000000000000c000 l5: 0000000000000001 l6: 0000000020000000 l7: ffff800047a0b734 i0: ffff800014feb000 i1: ffff800047a0b730 i2: 0000000000000001 i3: 000000000000000d i4: 0000000000000000 i5: 0000000000000000 i6: ffff800047a0ae81 i7: 00000000101888b0 I7: <niu_try_msix.constprop.0+0xc0/0x130 [niu]> Call Trace: [<00000000101888b0>] niu_try_msix.constprop.0+0xc0/0x130 [niu] [<000000001018f840>] niu_get_invariants+0x183c/0x207c [niu] [<00000000101902fc>] niu_pci_init_one+0x27c/0x2fc [niu] [<00000000005ef3e4>] local_pci_probe+0x28/0x74 [<0000000000469240>] work_for_cpu_fn+0x8/0x1c [<000000000046b008>] process_scheduled_works+0x144/0x210 [<000000000046b518>] worker_thread+0x13c/0x1c0 [<00000000004710e0>] kthread+0xb8/0xc8 [<00000000004060c8>] ret_from_fork+0x1c/0x2c [<0000000000000000>] 0x0 Kernel panic - not syncing: Non-resumable error. Fixes: 7d5ec3d36123 ("PCI/MSI: Mask all unused MSI-X entries") Cc: stable(a)vger.kernel.org Signed-off-by: Jonathan Currier <dullfire(a)yahoo.com> --- drivers/net/ethernet/sun/niu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/sun/niu.c b/drivers/net/ethernet/sun/niu.c index 41a27ae58ced..f5449b73b9a7 100644 --- a/drivers/net/ethernet/sun/niu.c +++ b/drivers/net/ethernet/sun/niu.c @@ -9058,6 +9058,8 @@ static void niu_try_msix(struct niu *np, u8 *ldg_num_map) msi_vec[i].entry = i; } + pdev->dev_flags |= PCI_DEV_FLAGS_MSIX_TOUCH_ENTRY_DATA_FIRST; + num_irqs = pci_enable_msix_range(pdev, msi_vec, 1, num_irqs); if (num_irqs < 0) { np->flags &= ~NIU_FLAGS_MSIX; -- 2.45.2

1 month, 4 weeks

2
1
0 0

[PATCH] EDAC/amd64: Fix size calculation for Non-Power-of-Two DIMMs

by Avadhut Naik

Each Chip-Select (CS) of a Unified Memory Controller (UMC) on AMD EPYC SOCs has an Address Mask and a Secondary Address Mask register associated with it. The amd64_edac module logs DIMM sizes on a per-UMC per-CS granularity during init using these two registers. Currently, the module primarily considers only the Address Mask register for computing DIMM sizes. The Secondary Address Mask register is only considered for odd CS. Additionally, if it has been considered, the Address Mask register is ignored altogether for that CS. For power-of-two DIMMs, this is not an issue since only the Address Mask register is used. For non-power-of-two DIMMs, however, the Secondary Address Mask register is used in conjunction with the Address Mask register. However, since the module only considers either of the two registers for a CS, the size computed by the module is incorrect. The Secondary Address Mask register is not considered for even CS, and the Address Mask register is not considered for odd CS. Furthermore, also rename some variables for greater clarity. Fixes: 81f5090db843 ("EDAC/amd64: Support asymmetric dual-rank DIMMs") Signed-off-by: Avadhut Naik <avadhut.naik(a)amd.com> Cc: stable(a)vger.kernel.org --- drivers/edac/amd64_edac.c | 56 ++++++++++++++++++++++++--------------- 1 file changed, 35 insertions(+), 21 deletions(-) diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c index 90f0eb7cc5b9..16117fda727f 100644 --- a/drivers/edac/amd64_edac.c +++ b/drivers/edac/amd64_edac.c @@ -1209,7 +1209,9 @@ static int umc_get_cs_mode(int dimm, u8 ctrl, struct amd64_pvt *pvt) if (csrow_enabled(2 * dimm + 1, ctrl, pvt)) cs_mode |= CS_ODD_PRIMARY; - /* Asymmetric dual-rank DIMM support. */ + if (csrow_sec_enabled(2 * dimm, ctrl, pvt)) + cs_mode |= CS_EVEN_SECONDARY; + if (csrow_sec_enabled(2 * dimm + 1, ctrl, pvt)) cs_mode |= CS_ODD_SECONDARY; @@ -1230,12 +1232,10 @@ static int umc_get_cs_mode(int dimm, u8 ctrl, struct amd64_pvt *pvt) return cs_mode; } -static int __addr_mask_to_cs_size(u32 addr_mask_orig, unsigned int cs_mode, - int csrow_nr, int dimm) +static int calculate_cs_size(u32 mask, unsigned int cs_mode) { - u32 msb, weight, num_zero_bits; - u32 addr_mask_deinterleaved; - int size = 0; + int msb, weight, num_zero_bits; + u32 deinterleaved_mask = 0; /* * The number of zero bits in the mask is equal to the number of bits @@ -1248,19 +1248,32 @@ static int __addr_mask_to_cs_size(u32 addr_mask_orig, unsigned int cs_mode, * without swapping with the most significant bit. This can be handled * by keeping the MSB where it is and ignoring the single zero bit. */ - msb = fls(addr_mask_orig) - 1; - weight = hweight_long(addr_mask_orig); + msb = fls(mask) - 1; + weight = hweight_long(mask); num_zero_bits = msb - weight - !!(cs_mode & CS_3R_INTERLEAVE); /* Take the number of zero bits off from the top of the mask. */ - addr_mask_deinterleaved = GENMASK_ULL(msb - num_zero_bits, 1); + deinterleaved_mask = GENMASK(msb - num_zero_bits, 1); + edac_dbg(1, " Deinterleaved AddrMask: 0x%x\n", deinterleaved_mask); + + return (deinterleaved_mask >> 2) + 1; +} + +static int __addr_mask_to_cs_size(u32 addr_mask, u32 addr_mask_sec, + unsigned int cs_mode, int csrow_nr, int dimm) +{ + int size = 0; edac_dbg(1, "CS%d DIMM%d AddrMasks:\n", csrow_nr, dimm); - edac_dbg(1, " Original AddrMask: 0x%x\n", addr_mask_orig); - edac_dbg(1, " Deinterleaved AddrMask: 0x%x\n", addr_mask_deinterleaved); + edac_dbg(1, " Primary AddrMask: 0x%x\n", addr_mask); /* Register [31:1] = Address [39:9]. Size is in kBs here. */ - size = (addr_mask_deinterleaved >> 2) + 1; + size = calculate_cs_size(addr_mask, cs_mode); + + if (addr_mask_sec) { + edac_dbg(1, " Secondary AddrMask: 0x%x\n", addr_mask); + size += calculate_cs_size(addr_mask_sec, cs_mode); + } /* Return size in MBs. */ return size >> 10; @@ -1270,7 +1283,7 @@ static int umc_addr_mask_to_cs_size(struct amd64_pvt *pvt, u8 umc, unsigned int cs_mode, int csrow_nr) { int cs_mask_nr = csrow_nr; - u32 addr_mask_orig; + u32 addr_mask = 0, addr_mask_sec = 0; int dimm, size = 0; /* No Chip Selects are enabled. */ @@ -1308,13 +1321,13 @@ static int umc_addr_mask_to_cs_size(struct amd64_pvt *pvt, u8 umc, if (!pvt->flags.zn_regs_v2) cs_mask_nr >>= 1; - /* Asymmetric dual-rank DIMM support. */ - if ((csrow_nr & 1) && (cs_mode & CS_ODD_SECONDARY)) - addr_mask_orig = pvt->csels[umc].csmasks_sec[cs_mask_nr]; - else - addr_mask_orig = pvt->csels[umc].csmasks[cs_mask_nr]; + if (cs_mode & CS_EVEN_PRIMARY || cs_mode & CS_ODD_PRIMARY) + addr_mask = pvt->csels[umc].csmasks[cs_mask_nr]; + + if (cs_mode & CS_EVEN_SECONDARY || cs_mode & CS_ODD_SECONDARY) + addr_mask_sec = pvt->csels[umc].csmasks_sec[cs_mask_nr]; - return __addr_mask_to_cs_size(addr_mask_orig, cs_mode, csrow_nr, dimm); + return __addr_mask_to_cs_size(addr_mask, addr_mask_sec, cs_mode, csrow_nr, dimm); } static void umc_debug_display_dimm_sizes(struct amd64_pvt *pvt, u8 ctrl) @@ -3512,9 +3525,10 @@ static void gpu_get_err_info(struct mce *m, struct err_info *err) static int gpu_addr_mask_to_cs_size(struct amd64_pvt *pvt, u8 umc, unsigned int cs_mode, int csrow_nr) { - u32 addr_mask_orig = pvt->csels[umc].csmasks[csrow_nr]; + u32 addr_mask = pvt->csels[umc].csmasks[csrow_nr]; + u32 addr_mask_sec = pvt->csels[umc].csmasks_sec[csrow_nr]; - return __addr_mask_to_cs_size(addr_mask_orig, cs_mode, csrow_nr, csrow_nr >> 1); + return __addr_mask_to_cs_size(addr_mask, addr_mask_sec, cs_mode, csrow_nr, csrow_nr >> 1); } static void gpu_debug_display_dimm_sizes(struct amd64_pvt *pvt, u8 ctrl) base-commit: f1861fb575b028e35e6233295441d535f2e3f240 -- 2.43.0

2 months

3
4
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror March 2025