November 2024 - Linux-stable-mirror

Re: [PATCH 6.6 000/213] 6.6.57-rc1 review

by Ivan Fay

Sent from my iPhone

5 months

1
0
0 0

[PATCH 2/2] drm/xe: improve hibernation on igpu

by Matthew Auld

The GGTT looks to be stored inside stolen memory on igpu which is not treated as normal RAM. The core kernel skips this memory range when creating the hibernation image, therefore when coming back from hibernation the GGTT programming is lost. This seems to cause issues with broken resume where GuC FW fails to load: [drm] *ERROR* GT0: load failed: status = 0x400000A0, time = 10ms, freq = 1250MHz (req 1300MHz), done = -1 [drm] *ERROR* GT0: load failed: status: Reset = 0, BootROM = 0x50, UKernel = 0x00, MIA = 0x00, Auth = 0x01 [drm] *ERROR* GT0: firmware signature verification failed [drm] *ERROR* CRITICAL: Xe has declared device 0000:00:02.0 as wedged. Current GGTT users are kernel internal and tracked as pinned, so it should be possible to hook into the existing save/restore logic that we use for dgpu, where the actual evict is skipped but on restore we importantly restore the GGTT programming. This has been confirmed to fix hibernation on at least ADL and MTL, though likely all igpu platforms are affected. This also means we have a hole in our testing, where the existing s4 tests only really test the driver hooks, and don't go as far as actually rebooting and restoring from the hibernation image and in turn powering down RAM (and therefore losing the contents of stolen). Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs") Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/3275 Signed-off-by: Matthew Auld <matthew.auld(a)intel.com> Cc: Matthew Brost <matthew.brost(a)intel.com> Cc: <stable(a)vger.kernel.org> # v6.8+ --- drivers/gpu/drm/xe/xe_bo.c | 36 ++++++++++++++------------------ drivers/gpu/drm/xe/xe_bo_evict.c | 6 ------ 2 files changed, 16 insertions(+), 26 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c index d79d8ef5c7d5..0ae5c8f7bab8 100644 --- a/drivers/gpu/drm/xe/xe_bo.c +++ b/drivers/gpu/drm/xe/xe_bo.c @@ -950,7 +950,10 @@ int xe_bo_restore_pinned(struct xe_bo *bo) if (WARN_ON(!xe_bo_is_pinned(bo))) return -EINVAL; - if (WARN_ON(xe_bo_is_vram(bo) || !bo->ttm.ttm)) + if (WARN_ON(xe_bo_is_vram(bo))) + return -EINVAL; + + if (WARN_ON(!bo->ttm.ttm && !xe_bo_is_stolen(bo))) return -EINVAL; if (!mem_type_is_vram(place->mem_type)) @@ -1770,6 +1773,7 @@ int xe_bo_pin_external(struct xe_bo *bo) int xe_bo_pin(struct xe_bo *bo) { + struct ttm_place *place = &(bo->placements[0]); struct xe_device *xe = xe_bo_device(bo); int err; @@ -1800,7 +1804,6 @@ int xe_bo_pin(struct xe_bo *bo) */ if (IS_DGFX(xe) && !(IS_ENABLED(CONFIG_DRM_XE_DEBUG) && bo->flags & XE_BO_FLAG_INTERNAL_TEST)) { - struct ttm_place *place = &(bo->placements[0]); if (mem_type_is_vram(place->mem_type)) { xe_assert(xe, place->flags & TTM_PL_FLAG_CONTIGUOUS); @@ -1809,13 +1812,12 @@ int xe_bo_pin(struct xe_bo *bo) vram_region_gpu_offset(bo->ttm.resource)) >> PAGE_SHIFT; place->lpfn = place->fpfn + (bo->size >> PAGE_SHIFT); } + } - if (mem_type_is_vram(place->mem_type) || - bo->flags & XE_BO_FLAG_GGTT) { - spin_lock(&xe->pinned.lock); - list_add_tail(&bo->pinned_link, &xe->pinned.kernel_bo_present); - spin_unlock(&xe->pinned.lock); - } + if (mem_type_is_vram(place->mem_type) || bo->flags & XE_BO_FLAG_GGTT) { + spin_lock(&xe->pinned.lock); + list_add_tail(&bo->pinned_link, &xe->pinned.kernel_bo_present); + spin_unlock(&xe->pinned.lock); } ttm_bo_pin(&bo->ttm); @@ -1863,24 +1865,18 @@ void xe_bo_unpin_external(struct xe_bo *bo) void xe_bo_unpin(struct xe_bo *bo) { + struct ttm_place *place = &(bo->placements[0]); struct xe_device *xe = xe_bo_device(bo); xe_assert(xe, !bo->ttm.base.import_attach); xe_assert(xe, xe_bo_is_pinned(bo)); - if (IS_DGFX(xe) && !(IS_ENABLED(CONFIG_DRM_XE_DEBUG) && - bo->flags & XE_BO_FLAG_INTERNAL_TEST)) { - struct ttm_place *place = &(bo->placements[0]); - - if (mem_type_is_vram(place->mem_type) || - bo->flags & XE_BO_FLAG_GGTT) { - spin_lock(&xe->pinned.lock); - xe_assert(xe, !list_empty(&bo->pinned_link)); - list_del_init(&bo->pinned_link); - spin_unlock(&xe->pinned.lock); - } + if (mem_type_is_vram(place->mem_type) || bo->flags & XE_BO_FLAG_GGTT) { + spin_lock(&xe->pinned.lock); + xe_assert(xe, !list_empty(&bo->pinned_link)); + list_del_init(&bo->pinned_link); + spin_unlock(&xe->pinned.lock); } - ttm_bo_unpin(&bo->ttm); } diff --git a/drivers/gpu/drm/xe/xe_bo_evict.c b/drivers/gpu/drm/xe/xe_bo_evict.c index 32043e1e5a86..b01bc20eb90b 100644 --- a/drivers/gpu/drm/xe/xe_bo_evict.c +++ b/drivers/gpu/drm/xe/xe_bo_evict.c @@ -34,9 +34,6 @@ int xe_bo_evict_all(struct xe_device *xe) u8 id; int ret; - if (!IS_DGFX(xe)) - return 0; - /* User memory */ for (mem_type = XE_PL_VRAM0; mem_type <= XE_PL_VRAM1; ++mem_type) { struct ttm_resource_manager *man = @@ -125,9 +122,6 @@ int xe_bo_restore_kernel(struct xe_device *xe) struct xe_bo *bo; int ret; - if (!IS_DGFX(xe)) - return 0; - spin_lock(&xe->pinned.lock); for (;;) { bo = list_first_entry_or_null(&xe->pinned.evicted, -- 2.47.0

5 months

2
1
0 0

Re: Patch "lib: alloc_tag_module_unload must wait for pending kfree_rcu calls" has been added to the 6.11-stable tree

by Suren Baghdasaryan

On Fri, Nov 1, 2024 at 6:58 AM Sasha Levin <sashal(a)kernel.org> wrote: > > This is a note to let you know that I've just added the patch titled > > lib: alloc_tag_module_unload must wait for pending kfree_rcu calls > > to the 6.11-stable tree which can be found at: > http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… > > The filename of the patch is: > lib-alloc_tag_module_unload-must-wait-for-pending-kf.patch > and it can be found in the queue-6.11 subdirectory. Thanks Sasha! Could you please double-check that the prerequisite patch https://lore.kernel.org/all/20241021171003.2907935-1-surenb@google.com/ was also picked up? I don't see it in the queue-6.11 directory. Without that patch this one will cause build errors, that's why I sent them as a patchset. Thanks, Suren. > > If you, or anyone else, feels it should not be added to the stable tree, > please let <stable(a)vger.kernel.org> know about it. > > > > commit 536dfe685ebd28b27ebfbc3d4b9168207b7e28a3 > Author: Florian Westphal <fw(a)strlen.de> > Date: Mon Oct 7 22:52:24 2024 +0200 > > lib: alloc_tag_module_unload must wait for pending kfree_rcu calls > > [ Upstream commit dc783ba4b9df3fb3e76e968b2cbeb9960069263c ] > > Ben Greear reports following splat: > ------------[ cut here ]------------ > net/netfilter/nf_nat_core.c:1114 module nf_nat func:nf_nat_register_fn has 256 allocated at module unload > WARNING: CPU: 1 PID: 10421 at lib/alloc_tag.c:168 alloc_tag_module_unload+0x22b/0x3f0 > Modules linked in: nf_nat(-) btrfs ufs qnx4 hfsplus hfs minix vfat msdos fat > ... > Hardware name: Default string Default string/SKYBAY, BIOS 5.12 08/04/2020 > RIP: 0010:alloc_tag_module_unload+0x22b/0x3f0 > codetag_unload_module+0x19b/0x2a0 > ? codetag_load_module+0x80/0x80 > > nf_nat module exit calls kfree_rcu on those addresses, but the free > operation is likely still pending by the time alloc_tag checks for leaks. > > Wait for outstanding kfree_rcu operations to complete before checking > resolves this warning. > > Reproducer: > unshare -n iptables-nft -t nat -A PREROUTING -p tcp > grep nf_nat /proc/allocinfo # will list 4 allocations > rmmod nft_chain_nat > rmmod nf_nat # will WARN. > > [akpm(a)linux-foundation.org: add comment] > Link: https://lkml.kernel.org/r/20241007205236.11847-1-fw@strlen.de > Fixes: a473573964e5 ("lib: code tagging module support") > Signed-off-by: Florian Westphal <fw(a)strlen.de> > Reported-by: Ben Greear <greearb(a)candelatech.com> > Closes: https://lore.kernel.org/netdev/bdaaef9d-4364-4171-b82b-bcfc12e207eb@candela… > Cc: Uladzislau Rezki <urezki(a)gmail.com> > Cc: Vlastimil Babka <vbabka(a)suse.cz> > Cc: Suren Baghdasaryan <surenb(a)google.com> > Cc: Kent Overstreet <kent.overstreet(a)linux.dev> > Cc: <stable(a)vger.kernel.org> > Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> > Signed-off-by: Sasha Levin <sashal(a)kernel.org> > > diff --git a/lib/codetag.c b/lib/codetag.c > index afa8a2d4f3173..d1fbbb7c2ec3d 100644 > --- a/lib/codetag.c > +++ b/lib/codetag.c > @@ -228,6 +228,9 @@ bool codetag_unload_module(struct module *mod) > if (!mod) > return true; > > + /* await any module's kfree_rcu() operations to complete */ > + kvfree_rcu_barrier(); > + > mutex_lock(&codetag_lock); > list_for_each_entry(cttype, &codetag_types, link) { > struct codetag_module *found = NULL;

5 months

1
0
0 0

[PATCH] nvme: rdma: Add check for queue in nvmet_rdma_cm_handler()

by George Rurikov

From: MrRurikov <grurikov(a)gmal.com> After having been assigned to a NULL value at rdma.c:1758, pointer 'queue' is passed as 1st parameter in call to function 'nvmet_rdma_queue_established' at rdma.c:1773, as 1st parameter in call to function 'nvmet_rdma_queue_disconnect' at rdma.c:1787 and as 2nd parameter in call to function 'nvmet_rdma_queue_connect_fail' at rdma.c:1800, where it is dereferenced. I understand, that driver is confident that the RDMA_CM_EVENT_CONNECT_REQUEST event will occur first and perform initialization, but maliciously prepared hardware could send events in violation of the protocol. Nothing guarantees that the sequence of events will start with RDMA_CM_EVENT_CONNECT_REQUEST. Found by Linux Verification Center (linuxtesting.org) with SVACE Fixes: e1a2ee249b19 ("nvmet-rdma: Fix use after free in nvmet_rdma_cm_handler()") Cc: stable(a)vger.kernel.org Signed-off-by: George Rurikov <g.ryurikov(a)securitycode.ru> --- drivers/nvme/target/rdma.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c index 1b6264fa5803..becebc95f349 100644 --- a/drivers/nvme/target/rdma.c +++ b/drivers/nvme/target/rdma.c @@ -1770,8 +1770,10 @@ static int nvmet_rdma_cm_handler(struct rdma_cm_id *cm_id, ret = nvmet_rdma_queue_connect(cm_id, event); break; case RDMA_CM_EVENT_ESTABLISHED: - nvmet_rdma_queue_established(queue); - break; + if (!queue) { + nvmet_rdma_queue_established(queue); + break; + } case RDMA_CM_EVENT_ADDR_CHANGE: if (!queue) { struct nvmet_rdma_port *port = cm_id->context; @@ -1782,8 +1784,10 @@ static int nvmet_rdma_cm_handler(struct rdma_cm_id *cm_id, fallthrough; case RDMA_CM_EVENT_DISCONNECTED: case RDMA_CM_EVENT_TIMEWAIT_EXIT: - nvmet_rdma_queue_disconnect(queue); - break; + if (!queue) { + nvmet_rdma_queue_disconnect(queue); + break; + } case RDMA_CM_EVENT_DEVICE_REMOVAL: ret = nvmet_rdma_device_removal(cm_id, queue); break; @@ -1793,8 +1797,10 @@ static int nvmet_rdma_cm_handler(struct rdma_cm_id *cm_id, fallthrough; case RDMA_CM_EVENT_UNREACHABLE: case RDMA_CM_EVENT_CONNECT_ERROR: - nvmet_rdma_queue_connect_fail(cm_id, queue); - break; + if (!queue) { + nvmet_rdma_queue_connect_fail(cm_id, queue); + break; + } default: pr_err("received unrecognized RDMA CM event %d\n", event->event); -- 2.34.1

5 months

3
3
0 0

Passenger traffic 2024

by Camille Batiste

Hey, Would you be interested in acquiring the attendees list of Passenger traffic Expo 2024? List contains: Names, Titles, Phone Numbers, Company Details, and more… Interested? Let me know so that I’ll send you the pricing for the same. Kind Regards, Camille Batiste Marketing Executive If you do not wish to receive our emails, please reply with "Not Interested."

5 months

1
0
0 0

[PATCH 0/1: 5.10/5.15] net: bridge: xmit: make sure we have at least eth header len bytes

by Randy.MacLeod＠windriver.com

From: Randy MacLeod <Randy.MacLeod(a)windriver.com> This is my first commit to -stable so I'm going to carefully explain what I've done. I work on the Yocto Project and I have done some work on the Linux network stack a long time ago so I'm not quite a complete newbie. I took the commit found here: https://lore.kernel.org/stable/20240527185645.658299380@linuxfoundation.org/ and backported as per my commit log: Based on above commit but simplified since pskb_may_pull_reason() does not exist until 6.1. I also trimmed the original commit log of the "Tested by dropwatch" section as well as the full stack trace since that may have changed in 5.10/5.15 and It compiles fine for 5.10 and 5.15 but I have not tested with dropwatch since the patch is just dropping short xmit packets for bridging. Finally, since the patch is much simpler than the original, I've removed the original patch author's SOB line. Please let me know if any of this is not what y'all'd like to see. Randy MacLeod (1): net: bridge: xmit: make sure we have at least eth header len bytes net/bridge/br_device.c | 5 +++++ 1 file changed, 5 insertions(+) base-commit: 5a8fa04b2a4de1d52be4a04690dcb52ac7998943 -- 2.34.1

5 months

1
1
0 0

[PATCH v2] drm/bridge: Fix assignment of the of_node of the parent to aux bridge

by Abel Vesa

The assignment of the of_node to the aux bridge needs to mark the of_node as reused as well, otherwise resource providers like pinctrl will report a gpio as already requested by a different device when both pinconf and gpios property are present. Fix that by using the device_set_of_node_from_dev() helper instead. Fixes: 6914968a0b52 ("drm/bridge: properly refcount DT nodes in aux bridge drivers") Cc: stable(a)vger.kernel.org # 6.8 Cc: Dmitry Baryshkov <dmitry.baryshkov(a)linaro.org> Signed-off-by: Abel Vesa <abel.vesa(a)linaro.org> --- Changes in v2: - Re-worded commit to be more explicit of what it fixes, as Johan suggested - Used device_set_of_node_from_dev() helper, as per Johan's suggestion - Added Fixes tag and cc'ed stable - Link to v1: https://lore.kernel.org/r/20241017-drm-aux-bridge-mark-of-node-reused-v1-1-… --- drivers/gpu/drm/bridge/aux-bridge.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/bridge/aux-bridge.c b/drivers/gpu/drm/bridge/aux-bridge.c index b29980f95379ec7af873ed6e0fb79a9abb663c7b..295e9d031e2dc86cbfd2a7350718fca181c99487 100644 --- a/drivers/gpu/drm/bridge/aux-bridge.c +++ b/drivers/gpu/drm/bridge/aux-bridge.c @@ -58,9 +58,10 @@ int drm_aux_bridge_register(struct device *parent) adev->id = ret; adev->name = "aux_bridge"; adev->dev.parent = parent; - adev->dev.of_node = of_node_get(parent->of_node); adev->dev.release = drm_aux_bridge_release; + device_set_of_node_from_dev(&adev->dev, parent); + ret = auxiliary_device_init(adev); if (ret) { ida_free(&drm_aux_bridge_ida, adev->id); --- base-commit: d61a00525464bfc5fe92c6ad713350988e492b88 change-id: 20241017-drm-aux-bridge-mark-of-node-reused-5c2ee740ff19 Best regards, -- Abel Vesa <abel.vesa(a)linaro.org>

5 months

6
25
0 0

[PATCH] vfio/qat: fix overflow check in qat_vf_resume_write()

by Giovanni Cabiddu

The unsigned variable `size_t len` is cast to the signed type `loff_t` when passed to the function check_add_overflow(). This function considers the type of the destination, which is of type loff_t (signed), potentially leading to an overflow. This issue is similar to the one described in the link below. Remove the cast. Note that even if check_add_overflow() is bypassed, by setting `len` to a value that is greater than LONG_MAX (which is considered as a negative value after the cast), the function copy_from_user(), invoked a few lines later, will not perform any copy and return `len` as (len > INT_MAX) causing qat_vf_resume_write() to fail with -EFAULT. Fixes: bb208810b1ab ("vfio/qat: Add vfio_pci driver for Intel QAT SR-IOV VF devices") CC: stable(a)vger.kernel.org # 6.10+ Link: https://lore.kernel.org/all/138bd2e2-ede8-4bcc-aa7b-f3d9de167a37@moroto.mou… Reported-by: Zijie Zhao <zzjas98(a)gmail.com> Signed-off-by: Giovanni Cabiddu <giovanni.cabiddu(a)intel.com> Reviewed-by: Xin Zeng <xin.zeng(a)intel.com> --- drivers/vfio/pci/qat/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/vfio/pci/qat/main.c b/drivers/vfio/pci/qat/main.c index e36740a282e7..1e3563fe7cab 100644 --- a/drivers/vfio/pci/qat/main.c +++ b/drivers/vfio/pci/qat/main.c @@ -305,7 +305,7 @@ static ssize_t qat_vf_resume_write(struct file *filp, const char __user *buf, offs = &filp->f_pos; if (*offs < 0 || - check_add_overflow((loff_t)len, *offs, &end)) + check_add_overflow(len, *offs, &end)) return -EOVERFLOW; if (end > mig_dev->state_size) -- 2.47.0

5 months

2
1
0 0

[PATCH v2 1/5] xhci: Combine two if statements for Etron xHCI host

by Kuangyi Chiang

Combine two if statements, because these hosts have the same quirk flags applied. Fixes: 91f7a1524a92 ("xhci: Apply broken streams quirk to Etron EJ188 xHCI host") Cc: <stable(a)vger.kernel.org> Signed-off-by: Kuangyi Chiang <ki.chiang65(a)gmail.com> --- drivers/usb/host/xhci-pci.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c index 7e538194a0a4..33a6d99afc10 100644 --- a/drivers/usb/host/xhci-pci.c +++ b/drivers/usb/host/xhci-pci.c @@ -395,12 +395,8 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci) xhci->quirks |= XHCI_DEFAULT_PM_RUNTIME_ALLOW; if (pdev->vendor == PCI_VENDOR_ID_ETRON && - pdev->device == PCI_DEVICE_ID_EJ168) { - xhci->quirks |= XHCI_RESET_ON_RESUME; - xhci->quirks |= XHCI_BROKEN_STREAMS; - } - if (pdev->vendor == PCI_VENDOR_ID_ETRON && - pdev->device == PCI_DEVICE_ID_EJ188) { + (pdev->device == PCI_DEVICE_ID_EJ168 || + pdev->device == PCI_DEVICE_ID_EJ188)) { xhci->quirks |= XHCI_RESET_ON_RESUME; xhci->quirks |= XHCI_BROKEN_STREAMS; } -- 2.25.1

5 months

2
3
0 0

[PATCH 6.6] rcu-tasks: Fix access non-existent percpu rtpcp variable in rcu_tasks_need_gpcb()

by liwei

From: Zqiang <qiang.zhang1211(a)gmail.com> [ Upstream commit fd70e9f1d85f5323096ad313ba73f5fe3d15ea41 ] For kernels built with CONFIG_FORCE_NR_CPUS=y, the nr_cpu_ids is defined as NR_CPUS instead of the number of possible cpus, this will cause the following system panic: smpboot: Allowing 4 CPUs, 0 hotplug CPUs ... setup_percpu: NR_CPUS:512 nr_cpumask_bits:512 nr_cpu_ids:512 nr_node_ids:1 ... BUG: unable to handle page fault for address: ffffffff9911c8c8 Oops: 0000 [#1] PREEMPT SMP PTI CPU: 0 PID: 15 Comm: rcu_tasks_trace Tainted: G W 6.6.21 #1 5dc7acf91a5e8e9ac9dcfc35bee0245691283ea6 RIP: 0010:rcu_tasks_need_gpcb+0x25d/0x2c0 RSP: 0018:ffffa371c00a3e60 EFLAGS: 00010082 CR2: ffffffff9911c8c8 CR3: 000000040fa20005 CR4: 00000000001706f0 Call Trace: <TASK> ? __die+0x23/0x80 ? page_fault_oops+0xa4/0x180 ? exc_page_fault+0x152/0x180 ? asm_exc_page_fault+0x26/0x40 ? rcu_tasks_need_gpcb+0x25d/0x2c0 ? __pfx_rcu_tasks_kthread+0x40/0x40 rcu_tasks_one_gp+0x69/0x180 rcu_tasks_kthread+0x94/0xc0 kthread+0xe8/0x140 ? __pfx_kthread+0x40/0x40 ret_from_fork+0x34/0x80 ? __pfx_kthread+0x40/0x40 ret_from_fork_asm+0x1b/0x80 </TASK> Considering that there may be holes in the CPU numbers, use the maximum possible cpu number, instead of nr_cpu_ids, for configuring enqueue and dequeue limits. [ neeraj.upadhyay: Fix htmldocs build error reported by Stephen Rothwell ] Closes: https://lore.kernel.org/linux-input/CALMA0xaTSMN+p4xUXkzrtR5r6k7hgoswcaXx7b… Reported-by: Zhixu Liu <zhixu.liu(a)gmail.com> Signed-off-by: Zqiang <qiang.zhang1211(a)gmail.com> Signed-off-by: Neeraj Upadhyay <neeraj.upadhyay(a)kernel.org> [liwei: adaptation context conflict] Signed-off-by: liwei <liwei728(a)huawei.com> --- kernel/rcu/tasks.h | 83 ++++++++++++++++++++++++++++++---------------- 1 file changed, 54 insertions(+), 29 deletions(-) diff --git a/kernel/rcu/tasks.h b/kernel/rcu/tasks.h index df81506cf2bd..fbfa0f90f884 100644 --- a/kernel/rcu/tasks.h +++ b/kernel/rcu/tasks.h @@ -33,6 +33,7 @@ typedef void (*postgp_func_t)(struct rcu_tasks *rtp); * @barrier_q_head: RCU callback for barrier operation. * @rtp_blkd_tasks: List of tasks blocked as readers. * @cpu: CPU number corresponding to this entry. + * @index: Index of this CPU in rtpcp_array of the rcu_tasks structure. * @rtpp: Pointer to the rcu_tasks structure. */ struct rcu_tasks_percpu { @@ -47,6 +48,7 @@ struct rcu_tasks_percpu { struct rcu_head barrier_q_head; struct list_head rtp_blkd_tasks; int cpu; + int index; struct rcu_tasks *rtpp; }; @@ -73,6 +75,7 @@ struct rcu_tasks_percpu { * @postgp_func: This flavor's post-grace-period function (optional). * @call_func: This flavor's call_rcu()-equivalent function. * @rtpcpu: This flavor's rcu_tasks_percpu structure. + * @rtpcp_array: Array of pointers to rcu_tasks_percpu structure of CPUs in cpu_possible_mask. * @percpu_enqueue_shift: Shift down CPU ID this much when enqueuing callbacks. * @percpu_enqueue_lim: Number of per-CPU callback queues in use for enqueuing. * @percpu_dequeue_lim: Number of per-CPU callback queues in use for dequeuing. @@ -106,6 +109,7 @@ struct rcu_tasks { postgp_func_t postgp_func; call_rcu_func_t call_func; struct rcu_tasks_percpu __percpu *rtpcpu; + struct rcu_tasks_percpu **rtpcp_array; int percpu_enqueue_shift; int percpu_enqueue_lim; int percpu_dequeue_lim; @@ -179,6 +183,8 @@ module_param(rcu_task_collapse_lim, int, 0444); static int rcu_task_lazy_lim __read_mostly = 32; module_param(rcu_task_lazy_lim, int, 0444); +static int rcu_task_cpu_ids; + /* RCU tasks grace-period state for debugging. */ #define RTGS_INIT 0 #define RTGS_WAIT_WAIT_CBS 1 @@ -243,6 +249,8 @@ static void cblist_init_generic(struct rcu_tasks *rtp) unsigned long flags; int lim; int shift; + int maxcpu; + int index = 0; if (rcu_task_enqueue_lim < 0) { rcu_task_enqueue_lim = 1; @@ -252,14 +260,9 @@ static void cblist_init_generic(struct rcu_tasks *rtp) } lim = rcu_task_enqueue_lim; - if (lim > nr_cpu_ids) - lim = nr_cpu_ids; - shift = ilog2(nr_cpu_ids / lim); - if (((nr_cpu_ids - 1) >> shift) >= lim) - shift++; - WRITE_ONCE(rtp->percpu_enqueue_shift, shift); - WRITE_ONCE(rtp->percpu_dequeue_lim, lim); - smp_store_release(&rtp->percpu_enqueue_lim, lim); + rtp->rtpcp_array = kcalloc(num_possible_cpus(), sizeof(struct rcu_tasks_percpu *), GFP_KERNEL); + BUG_ON(!rtp->rtpcp_array); + for_each_possible_cpu(cpu) { struct rcu_tasks_percpu *rtpcp = per_cpu_ptr(rtp->rtpcpu, cpu); @@ -273,12 +276,28 @@ static void cblist_init_generic(struct rcu_tasks *rtp) INIT_WORK(&rtpcp->rtp_work, rcu_tasks_invoke_cbs_wq); rtpcp->cpu = cpu; rtpcp->rtpp = rtp; + rtpcp->index = index; + rtp->rtpcp_array[index] = rtpcp; + index++; if (!rtpcp->rtp_blkd_tasks.next) INIT_LIST_HEAD(&rtpcp->rtp_blkd_tasks); + + maxcpu = cpu; } - pr_info("%s: Setting shift to %d and lim to %d rcu_task_cb_adjust=%d.\n", rtp->name, - data_race(rtp->percpu_enqueue_shift), data_race(rtp->percpu_enqueue_lim), rcu_task_cb_adjust); + rcu_task_cpu_ids = maxcpu + 1; + if (lim > rcu_task_cpu_ids) + lim = rcu_task_cpu_ids; + shift = ilog2(rcu_task_cpu_ids / lim); + if (((rcu_task_cpu_ids - 1) >> shift) >= lim) + shift++; + WRITE_ONCE(rtp->percpu_enqueue_shift, shift); + WRITE_ONCE(rtp->percpu_dequeue_lim, lim); + smp_store_release(&rtp->percpu_enqueue_lim, lim); + + pr_info("%s: Setting shift to %d and lim to %d rcu_task_cb_adjust=%d rcu_task_cpu_ids=%d.\n", + rtp->name, data_race(rtp->percpu_enqueue_shift), data_race(rtp->percpu_enqueue_lim), + rcu_task_cb_adjust, rcu_task_cpu_ids); } // Compute wakeup time for lazy callback timer. @@ -346,7 +365,7 @@ static void call_rcu_tasks_generic(struct rcu_head *rhp, rcu_callback_t func, rtpcp->rtp_n_lock_retries = 0; } if (rcu_task_cb_adjust && ++rtpcp->rtp_n_lock_retries > rcu_task_contend_lim && - READ_ONCE(rtp->percpu_enqueue_lim) != nr_cpu_ids) + READ_ONCE(rtp->percpu_enqueue_lim) != rcu_task_cpu_ids) needadjust = true; // Defer adjustment to avoid deadlock. } // Queuing callbacks before initialization not yet supported. @@ -366,10 +385,10 @@ static void call_rcu_tasks_generic(struct rcu_head *rhp, rcu_callback_t func, raw_spin_unlock_irqrestore_rcu_node(rtpcp, flags); if (unlikely(needadjust)) { raw_spin_lock_irqsave(&rtp->cbs_gbl_lock, flags); - if (rtp->percpu_enqueue_lim != nr_cpu_ids) { + if (rtp->percpu_enqueue_lim != rcu_task_cpu_ids) { WRITE_ONCE(rtp->percpu_enqueue_shift, 0); - WRITE_ONCE(rtp->percpu_dequeue_lim, nr_cpu_ids); - smp_store_release(&rtp->percpu_enqueue_lim, nr_cpu_ids); + WRITE_ONCE(rtp->percpu_dequeue_lim, rcu_task_cpu_ids); + smp_store_release(&rtp->percpu_enqueue_lim, rcu_task_cpu_ids); pr_info("Switching %s to per-CPU callback queuing.\n", rtp->name); } raw_spin_unlock_irqrestore(&rtp->cbs_gbl_lock, flags); @@ -440,6 +459,8 @@ static int rcu_tasks_need_gpcb(struct rcu_tasks *rtp) int needgpcb = 0; for (cpu = 0; cpu < smp_load_acquire(&rtp->percpu_dequeue_lim); cpu++) { + if (!cpu_possible(cpu)) + continue; struct rcu_tasks_percpu *rtpcp = per_cpu_ptr(rtp->rtpcpu, cpu); /* Advance and accelerate any new callbacks. */ @@ -477,7 +498,7 @@ static int rcu_tasks_need_gpcb(struct rcu_tasks *rtp) if (rcu_task_cb_adjust && ncbs <= rcu_task_collapse_lim) { raw_spin_lock_irqsave(&rtp->cbs_gbl_lock, flags); if (rtp->percpu_enqueue_lim > 1) { - WRITE_ONCE(rtp->percpu_enqueue_shift, order_base_2(nr_cpu_ids)); + WRITE_ONCE(rtp->percpu_enqueue_shift, order_base_2(rcu_task_cpu_ids)); smp_store_release(&rtp->percpu_enqueue_lim, 1); rtp->percpu_dequeue_gpseq = get_state_synchronize_rcu(); gpdone = false; @@ -492,7 +513,9 @@ static int rcu_tasks_need_gpcb(struct rcu_tasks *rtp) pr_info("Completing switch %s to CPU-0 callback queuing.\n", rtp->name); } if (rtp->percpu_dequeue_lim == 1) { - for (cpu = rtp->percpu_dequeue_lim; cpu < nr_cpu_ids; cpu++) { + for (cpu = rtp->percpu_dequeue_lim; cpu < rcu_task_cpu_ids; cpu++) { + if (!cpu_possible(cpu)) + continue; struct rcu_tasks_percpu *rtpcp = per_cpu_ptr(rtp->rtpcpu, cpu); WARN_ON_ONCE(rcu_segcblist_n_cbs(&rtpcp->cblist)); @@ -507,30 +530,32 @@ static int rcu_tasks_need_gpcb(struct rcu_tasks *rtp) // Advance callbacks and invoke any that are ready. static void rcu_tasks_invoke_cbs(struct rcu_tasks *rtp, struct rcu_tasks_percpu *rtpcp) { - int cpu; - int cpunext; int cpuwq; unsigned long flags; int len; + int index; struct rcu_head *rhp; struct rcu_cblist rcl = RCU_CBLIST_INITIALIZER(rcl); struct rcu_tasks_percpu *rtpcp_next; - cpu = rtpcp->cpu; - cpunext = cpu * 2 + 1; - if (cpunext < smp_load_acquire(&rtp->percpu_dequeue_lim)) { - rtpcp_next = per_cpu_ptr(rtp->rtpcpu, cpunext); - cpuwq = rcu_cpu_beenfullyonline(cpunext) ? cpunext : WORK_CPU_UNBOUND; - queue_work_on(cpuwq, system_wq, &rtpcp_next->rtp_work); - cpunext++; - if (cpunext < smp_load_acquire(&rtp->percpu_dequeue_lim)) { - rtpcp_next = per_cpu_ptr(rtp->rtpcpu, cpunext); - cpuwq = rcu_cpu_beenfullyonline(cpunext) ? cpunext : WORK_CPU_UNBOUND; + index = rtpcp->index * 2 + 1; + if (index < num_possible_cpus()) { + rtpcp_next = rtp->rtpcp_array[index]; + if (rtpcp_next->cpu < smp_load_acquire(&rtp->percpu_dequeue_lim)) { + cpuwq = rcu_cpu_beenfullyonline(rtpcp_next->cpu) ? rtpcp_next->cpu : WORK_CPU_UNBOUND; queue_work_on(cpuwq, system_wq, &rtpcp_next->rtp_work); + index++; + if (index < num_possible_cpus()) { + rtpcp_next = rtp->rtpcp_array[index]; + if (rtpcp_next->cpu < smp_load_acquire(&rtp->percpu_dequeue_lim)) { + cpuwq = rcu_cpu_beenfullyonline(rtpcp_next->cpu) ? rtpcp_next->cpu : WORK_CPU_UNBOUND; + queue_work_on(cpuwq, system_wq, &rtpcp_next->rtp_work); + } + } } } - if (rcu_segcblist_empty(&rtpcp->cblist) || !cpu_possible(cpu)) + if (rcu_segcblist_empty(&rtpcp->cblist)) return; raw_spin_lock_irqsave_rcu_node(rtpcp, flags); rcu_segcblist_advance(&rtpcp->cblist, rcu_seq_current(&rtp->tasks_gp_seq)); -- 2.25.1

5 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror November 2024