December 2024 - Linux-stable-mirror

FAILED: patch "[PATCH] xen: Fix the issue of resource not being properly released in" failed to apply to 4.19-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.19-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-4.19.y git checkout FETCH_HEAD git cherry-pick -x afc545da381ba0c651b2658966ac737032676f01 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2024120221-serving-certainly-75b1@gregkh' --subject-prefix 'PATCH 4.19.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From afc545da381ba0c651b2658966ac737032676f01 Mon Sep 17 00:00:00 2001 From: Qiu-ji Chen <chenqiuji666(a)gmail.com> Date: Tue, 5 Nov 2024 21:09:19 +0800 Subject: [PATCH] xen: Fix the issue of resource not being properly released in xenbus_dev_probe() This patch fixes an issue in the function xenbus_dev_probe(). In the xenbus_dev_probe() function, within the if (err) branch at line 313, the program incorrectly returns err directly without releasing the resources allocated by err = drv->probe(dev, id). As the return value is non-zero, the upper layers assume the processing logic has failed. However, the probe operation was performed earlier without a corresponding remove operation. Since the probe actually allocates resources, failing to perform the remove operation could lead to problems. To fix this issue, we followed the resource release logic of the xenbus_dev_remove() function by adding a new block fail_remove before the fail_put block. After entering the branch if (err) at line 313, the function will use a goto statement to jump to the fail_remove block, ensuring that the previously acquired resources are correctly released, thus preventing the reference count leak. This bug was identified by an experimental static analysis tool developed by our team. The tool specializes in analyzing reference count operations and detecting potential issues where resources are not properly managed. In this case, the tool flagged the missing release operation as a potential problem, which led to the development of this patch. Fixes: 4bac07c993d0 ("xen: add the Xenbus sysfs and virtual device hotplug driver") Cc: stable(a)vger.kernel.org Signed-off-by: Qiu-ji Chen <chenqiuji666(a)gmail.com> Reviewed-by: Juergen Gross <jgross(a)suse.com> Message-ID: <20241105130919.4621-1-chenqiuji666(a)gmail.com> Signed-off-by: Juergen Gross <jgross(a)suse.com> diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c index 9f097f1f4a4c..6d32ffb01136 100644 --- a/drivers/xen/xenbus/xenbus_probe.c +++ b/drivers/xen/xenbus/xenbus_probe.c @@ -313,7 +313,7 @@ int xenbus_dev_probe(struct device *_dev) if (err) { dev_warn(&dev->dev, "watch_otherend on %s failed.\n", dev->nodename); - return err; + goto fail_remove; } dev->spurious_threshold = 1; @@ -322,6 +322,12 @@ int xenbus_dev_probe(struct device *_dev) dev->nodename); return 0; +fail_remove: + if (drv->remove) { + down(&dev->reclaim_sem); + drv->remove(dev); + up(&dev->reclaim_sem); + } fail_put: module_put(drv->driver.owner); fail:

1 month, 3 weeks

1
0
0 0

FAILED: patch "[PATCH] xen: Fix the issue of resource not being properly released in" failed to apply to 5.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.4-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-5.4.y git checkout FETCH_HEAD git cherry-pick -x afc545da381ba0c651b2658966ac737032676f01 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2024120220-amuck-esophagus-6542@gregkh' --subject-prefix 'PATCH 5.4.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From afc545da381ba0c651b2658966ac737032676f01 Mon Sep 17 00:00:00 2001 From: Qiu-ji Chen <chenqiuji666(a)gmail.com> Date: Tue, 5 Nov 2024 21:09:19 +0800 Subject: [PATCH] xen: Fix the issue of resource not being properly released in xenbus_dev_probe() This patch fixes an issue in the function xenbus_dev_probe(). In the xenbus_dev_probe() function, within the if (err) branch at line 313, the program incorrectly returns err directly without releasing the resources allocated by err = drv->probe(dev, id). As the return value is non-zero, the upper layers assume the processing logic has failed. However, the probe operation was performed earlier without a corresponding remove operation. Since the probe actually allocates resources, failing to perform the remove operation could lead to problems. To fix this issue, we followed the resource release logic of the xenbus_dev_remove() function by adding a new block fail_remove before the fail_put block. After entering the branch if (err) at line 313, the function will use a goto statement to jump to the fail_remove block, ensuring that the previously acquired resources are correctly released, thus preventing the reference count leak. This bug was identified by an experimental static analysis tool developed by our team. The tool specializes in analyzing reference count operations and detecting potential issues where resources are not properly managed. In this case, the tool flagged the missing release operation as a potential problem, which led to the development of this patch. Fixes: 4bac07c993d0 ("xen: add the Xenbus sysfs and virtual device hotplug driver") Cc: stable(a)vger.kernel.org Signed-off-by: Qiu-ji Chen <chenqiuji666(a)gmail.com> Reviewed-by: Juergen Gross <jgross(a)suse.com> Message-ID: <20241105130919.4621-1-chenqiuji666(a)gmail.com> Signed-off-by: Juergen Gross <jgross(a)suse.com> diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c index 9f097f1f4a4c..6d32ffb01136 100644 --- a/drivers/xen/xenbus/xenbus_probe.c +++ b/drivers/xen/xenbus/xenbus_probe.c @@ -313,7 +313,7 @@ int xenbus_dev_probe(struct device *_dev) if (err) { dev_warn(&dev->dev, "watch_otherend on %s failed.\n", dev->nodename); - return err; + goto fail_remove; } dev->spurious_threshold = 1; @@ -322,6 +322,12 @@ int xenbus_dev_probe(struct device *_dev) dev->nodename); return 0; +fail_remove: + if (drv->remove) { + down(&dev->reclaim_sem); + drv->remove(dev); + up(&dev->reclaim_sem); + } fail_put: module_put(drv->driver.owner); fail:

1 month, 3 weeks

1
0
0 0

5.10.225 stable kernel cgroup_mutex not held assertion failure

by Greg Thelen

Linux stable v5.10.226 suffers a lockdep warning when accessing /proc/PID/cpuset. cset_cgroup_from_root() is called without cgroup_mutex is held, which causes assertion failure. Bisect blames 5.10.225 commit 688325078a8b ("cgroup/cpuset: Prevent UAF in proc_cpuset_show()"). I've have not easily reproduced the problem that this change fixes, so I'm not sure if it's best to revert the fix or adapt it to meet the 5.10 locking expectations. The lockdep complaint: $ cat /proc/1/cpuset $ dmesg [ 198.744891] ------------[ cut here ]------------ [ 198.744918] WARNING: CPU: 4 PID: 9301 at kernel/cgroup/cgroup.c:1395 cset_cgroup_from_root+0xb2/0xd0 [ 198.744957] RIP: 0010:cset_cgroup_from_root+0xb2/0xd0 [ 198.744960] Code: 02 00 00 74 11 48 8b 09 48 39 cb 75 eb eb 19 49 83 c6 10 4c 89 f0 48 85 c0 74 0d 5b 41 5e c3 48 8b 43 60 48 85 c0 75 f3 0f 0b <0f> 0b 83 3d 69 01 ee 01 00 0f 85 78 ff ff ff eb 8b 0f 0b eb 87 66 [ 198.744962] RSP: 0018:ffffb492608a7ce8 EFLAGS: 00010046 [ 198.744977] RAX: 0000000000000000 RBX: ffffffff8f4171b8 RCX: cc949de848c33e00 [ 198.744979] RDX: 0000000000001000 RSI: ffffffff8f415450 RDI: ffff92e5417c4dc0 [ 198.744981] RBP: ffff9303467e3f00 R08: 0000000000000008 R09: ffffffff9122d568 [ 198.744983] R10: ffff92e5417c4380 R11: 0000000000000000 R12: ffff92e3d9506000 [ 198.744984] R13: 0000000000000000 R14: ffff92e443a96000 R15: ffff92e3d9506000 [ 198.744987] FS: 00007f15d94ed740(0000) GS:ffff9302bf500000(0000) knlGS:0000000000000000 [ 198.744988] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 198.744990] CR2: 00007f15d94ca000 CR3: 00000002816ca003 CR4: 00000000001706e0 [ 198.744992] Call Trace: [ 198.744996] ? __warn+0xcd/0x1c0 [ 198.745000] ? cset_cgroup_from_root+0xb2/0xd0 [ 198.745008] ? report_bug+0x87/0xf0 [ 198.745015] ? handle_bug+0x42/0x80 [ 198.745017] ? exc_invalid_op+0x16/0x70 [ 198.745021] ? asm_exc_invalid_op+0x12/0x20 [ 198.745030] ? cset_cgroup_from_root+0xb2/0xd0 [ 198.745034] ? cset_cgroup_from_root+0x28/0xd0 [ 198.745038] cgroup_path_ns_locked+0x23/0x50 [ 198.745044] proc_cpuset_show+0x115/0x210 [ 198.745049] proc_single_show+0x4a/0xa0 [ 198.745056] seq_read_iter+0x14d/0x400 [ 198.745063] seq_read+0x103/0x130 [ 198.745074] vfs_read+0xea/0x320 [ 198.745078] ? do_user_addr_fault+0x25b/0x390 [ 198.745085] ? do_user_addr_fault+0x25b/0x390 [ 198.745090] ksys_read+0x70/0xe0 [ 198.745096] do_syscall_64+0x2d/0x40 [ 198.745099] entry_SYSCALL_64_after_hwframe+0x61/0xcb

1 month, 3 weeks

8
28
0 0

[PATCH v2] arm64: dts: rockchip: increase gmac rx_delay to 0x11 on rk3399-puma

by Jakob Unterwurzacher

During mass manufacturing, we noticed the mmc_rx_crc_error counter, as reported by "ethtool -S eth0 | grep mmc_rx_crc_error" to increase above zero during nuttcp speedtests. Cycling through the rx_delay range on two boards shows that there is a large "good" region from 0x11 to 0x35 (see below for details). This commit increases rx_delay to 0x11, which is the smallest possible change that fixes the issue we are seeing on the KSZ9031 PHY. This also matches what most other rk3399 boards do. Tests for Puma PCBA S/N TT0069903: rx_delay mmc_rx_crc_error -------- ---------------- 0x09 (dhcp broken) 0x10 897 0x11 0 0x20 0 0x30 0 0x35 0 0x3a 745 0x3b 11375 0x3c 36680 0x40 (dhcp broken) 0x7f (dhcp broken) Tests for Puma PCBA S/N TT0157733: rx_delay mmc_rx_crc_error -------- ---------------- 0x10 59 0x11 0 0x35 0 Fixes: 2c66fc34e945 ("arm64: dts: rockchip: add RK3399-Q7 (Puma) SoM") Cc: <stable(a)vger.kernel.org> Reviewed-by: Quentin Schulz <quentin.schulz(a)cherry.de> Signed-off-by: Jakob Unterwurzacher <jakob.unterwurzacher(a)cherry.de> --- v2: cc stable, add "Fixes:", add omitted "there" to commit msg, add Reviewed-by. arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi index 9efcdce0f593..13d0c511046b 100644 --- a/arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi @@ -181,7 +181,7 @@ &gmac { snps,reset-active-low; snps,reset-delays-us = <0 10000 50000>; tx_delay = <0x10>; - rx_delay = <0x10>; + rx_delay = <0x11>; status = "okay"; }; -- 2.39.5

1 month, 3 weeks

1
0
0 0

[PATCH 6.6 0/3] Fix PPS channel routing

by Csókás, Bence

On certain i.MX8 series parts [1], the PPS channel 0 is routed internally to eDMA, and the external PPS pin is available on channel 1. In addition, on certain boards, the PPS may be wired on the PCB to an EVENTOUTn pin other than 0. On these systems it is necessary that the PPS channel be able to be configured from the Device Tree. [1] https://lore.kernel.org/all/ZrPYOWA3FESx197L@lizhi-Precision-Tower-5810/ Francesco Dolcini (3): dt-bindings: net: fec: add pps channel property net: fec: refactor PPS channel configuration net: fec: make PPS channel configurable Documentation/devicetree/bindings/net/fsl,fec.yaml | 7 +++++++ drivers/net/ethernet/freescale/fec_ptp.c | 11 ++++++----- 2 files changed, 13 insertions(+), 5 deletions(-) -- 2.34.1

1 month, 3 weeks

4
8
0 0

[PATCH] xhci: dbc: Fix STALL transfer event handling

by Mathias Nyman

commit 9044ad57b60b0556d42b6f8aa218a68865e810a4 upstream Backport targeted specifically for linux-6.6.y stable kernel. Resolve minor conflict due to 10-patch dbc cleanup series in 6.8 Don't flush all pending DbC data requests when an endpoint halts. An endpoint may halt and xHC DbC triggers a STALL error event if there's an issue with a bulk data transfer. The transfer should restart once xHC DbC receives a ClearFeature(ENDPOINT_HALT) request from the host. Once xHC DbC restarts it will start from the TRB pointed to by dequeue field in the endpoint context, which might be the same TRB we got the STALL event for. Turn the TRB to a no-op in this case to make sure xHC DbC doesn't reuse and tries to retransmit this same TRB after we already handled it, and gave its corresponding data request back. Other STALL events might be completely bogus. Lukasz Bartosik discovered that xHC DbC might issue spurious STALL events if hosts sends a ClearFeature(ENDPOINT_HALT) request to non-halted endpoints even without any active bulk transfers. Assume STALL event is spurious if it reports 0 bytes transferred, and the endpoint stopped on the STALLED TRB. Don't give back the data request corresponding to the TRB in this case. The halted status is per endpoint. Track it with a per endpoint flag instead of the driver invented DbC wide DS_STALLED state. DbC remains in DbC-Configured state even if endpoints halt. There is no Stalled state in the DbC Port state Machine (xhci section 7.6.6) Reported-by: Łukasz Bartosik <ukaszb(a)chromium.org> Closes: https://lore.kernel.org/linux-usb/20240725074857.623299-1-ukaszb@chromium.o… Tested-by: Łukasz Bartosik <ukaszb(a)chromium.org> Signed-off-by: Mathias Nyman <mathias.nyman(a)linux.intel.com> --- drivers/usb/host/xhci-dbgcap.c | 132 ++++++++++++++++++++------------- drivers/usb/host/xhci-dbgcap.h | 2 +- 2 files changed, 83 insertions(+), 51 deletions(-) diff --git a/drivers/usb/host/xhci-dbgcap.c b/drivers/usb/host/xhci-dbgcap.c index b40d9238d447..69067015f0d5 100644 --- a/drivers/usb/host/xhci-dbgcap.c +++ b/drivers/usb/host/xhci-dbgcap.c @@ -158,16 +158,18 @@ static void xhci_dbc_giveback(struct dbc_request *req, int status) spin_lock(&dbc->lock); } -static void xhci_dbc_flush_single_request(struct dbc_request *req) +static void trb_to_noop(union xhci_trb *trb) { - union xhci_trb *trb = req->trb; - trb->generic.field[0] = 0; trb->generic.field[1] = 0; trb->generic.field[2] = 0; trb->generic.field[3] &= cpu_to_le32(TRB_CYCLE); trb->generic.field[3] |= cpu_to_le32(TRB_TYPE(TRB_TR_NOOP)); +} +static void xhci_dbc_flush_single_request(struct dbc_request *req) +{ + trb_to_noop(req->trb); xhci_dbc_giveback(req, -ESHUTDOWN); } @@ -637,7 +639,6 @@ static void xhci_dbc_stop(struct xhci_dbc *dbc) case DS_DISABLED: return; case DS_CONFIGURED: - case DS_STALLED: if (dbc->driver->disconnect) dbc->driver->disconnect(dbc); break; @@ -657,6 +658,23 @@ static void xhci_dbc_stop(struct xhci_dbc *dbc) } } +static void +handle_ep_halt_changes(struct xhci_dbc *dbc, struct dbc_ep *dep, bool halted) +{ + if (halted) { + dev_info(dbc->dev, "DbC Endpoint halted\n"); + dep->halted = 1; + + } else if (dep->halted) { + dev_info(dbc->dev, "DbC Endpoint halt cleared\n"); + dep->halted = 0; + + if (!list_empty(&dep->list_pending)) + writel(DBC_DOOR_BELL_TARGET(dep->direction), + &dbc->regs->doorbell); + } +} + static void dbc_handle_port_status(struct xhci_dbc *dbc, union xhci_trb *event) { @@ -685,6 +703,7 @@ static void dbc_handle_xfer_event(struct xhci_dbc *dbc, union xhci_trb *event) struct xhci_ring *ring; int ep_id; int status; + struct xhci_ep_ctx *ep_ctx; u32 comp_code; size_t remain_length; struct dbc_request *req = NULL, *r; @@ -694,8 +713,30 @@ static void dbc_handle_xfer_event(struct xhci_dbc *dbc, union xhci_trb *event) ep_id = TRB_TO_EP_ID(le32_to_cpu(event->generic.field[3])); dep = (ep_id == EPID_OUT) ? get_out_ep(dbc) : get_in_ep(dbc); + ep_ctx = (ep_id == EPID_OUT) ? + dbc_bulkout_ctx(dbc) : dbc_bulkin_ctx(dbc); ring = dep->ring; + /* Match the pending request: */ + list_for_each_entry(r, &dep->list_pending, list_pending) { + if (r->trb_dma == event->trans_event.buffer) { + req = r; + break; + } + if (r->status == -COMP_STALL_ERROR) { + dev_warn(dbc->dev, "Give back stale stalled req\n"); + ring->num_trbs_free++; + xhci_dbc_giveback(r, 0); + } + } + + if (!req) { + dev_warn(dbc->dev, "no matched request\n"); + return; + } + + trace_xhci_dbc_handle_transfer(ring, &req->trb->generic); + switch (comp_code) { case COMP_SUCCESS: remain_length = 0; @@ -706,31 +747,49 @@ static void dbc_handle_xfer_event(struct xhci_dbc *dbc, union xhci_trb *event) case COMP_TRB_ERROR: case COMP_BABBLE_DETECTED_ERROR: case COMP_USB_TRANSACTION_ERROR: - case COMP_STALL_ERROR: dev_warn(dbc->dev, "tx error %d detected\n", comp_code); status = -comp_code; break; + case COMP_STALL_ERROR: + dev_warn(dbc->dev, "Stall error at bulk TRB %llx, remaining %zu, ep deq %llx\n", + event->trans_event.buffer, remain_length, ep_ctx->deq); + status = 0; + dep->halted = 1; + + /* + * xHC DbC may trigger a STALL bulk xfer event when host sends a + * ClearFeature(ENDPOINT_HALT) request even if there wasn't an + * active bulk transfer. + * + * Don't give back this transfer request as hardware will later + * start processing TRBs starting from this 'STALLED' TRB, + * causing TRBs and requests to be out of sync. + * + * If STALL event shows some bytes were transferred then assume + * it's an actual transfer issue and give back the request. + * In this case mark the TRB as No-Op to avoid hw from using the + * TRB again. + */ + + if ((ep_ctx->deq & ~TRB_CYCLE) == event->trans_event.buffer) { + dev_dbg(dbc->dev, "Ep stopped on Stalled TRB\n"); + if (remain_length == req->length) { + dev_dbg(dbc->dev, "Spurious stall event, keep req\n"); + req->status = -COMP_STALL_ERROR; + req->actual = 0; + return; + } + dev_dbg(dbc->dev, "Give back stalled req, but turn TRB to No-op\n"); + trb_to_noop(req->trb); + } + break; + default: dev_err(dbc->dev, "unknown tx error %d\n", comp_code); status = -comp_code; break; } - /* Match the pending request: */ - list_for_each_entry(r, &dep->list_pending, list_pending) { - if (r->trb_dma == event->trans_event.buffer) { - req = r; - break; - } - } - - if (!req) { - dev_warn(dbc->dev, "no matched request\n"); - return; - } - - trace_xhci_dbc_handle_transfer(ring, &req->trb->generic); - ring->num_trbs_free++; req->actual = req->length - remain_length; xhci_dbc_giveback(req, status); @@ -750,7 +809,6 @@ static void inc_evt_deq(struct xhci_ring *ring) static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) { dma_addr_t deq; - struct dbc_ep *dep; union xhci_trb *evt; u32 ctrl, portsc; bool update_erdp = false; @@ -802,43 +860,17 @@ static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) return EVT_DISC; } - /* Handle endpoint stall event: */ + /* Check and handle changes in endpoint halt status */ ctrl = readl(&dbc->regs->control); - if ((ctrl & DBC_CTRL_HALT_IN_TR) || - (ctrl & DBC_CTRL_HALT_OUT_TR)) { - dev_info(dbc->dev, "DbC Endpoint stall\n"); - dbc->state = DS_STALLED; - - if (ctrl & DBC_CTRL_HALT_IN_TR) { - dep = get_in_ep(dbc); - xhci_dbc_flush_endpoint_requests(dep); - } - - if (ctrl & DBC_CTRL_HALT_OUT_TR) { - dep = get_out_ep(dbc); - xhci_dbc_flush_endpoint_requests(dep); - } - - return EVT_DONE; - } + handle_ep_halt_changes(dbc, get_in_ep(dbc), ctrl & DBC_CTRL_HALT_IN_TR); + handle_ep_halt_changes(dbc, get_out_ep(dbc), ctrl & DBC_CTRL_HALT_OUT_TR); /* Clear DbC run change bit: */ if (ctrl & DBC_CTRL_DBC_RUN_CHANGE) { writel(ctrl, &dbc->regs->control); ctrl = readl(&dbc->regs->control); } - break; - case DS_STALLED: - ctrl = readl(&dbc->regs->control); - if (!(ctrl & DBC_CTRL_HALT_IN_TR) && - !(ctrl & DBC_CTRL_HALT_OUT_TR) && - (ctrl & DBC_CTRL_DBC_RUN)) { - dbc->state = DS_CONFIGURED; - break; - } - - return EVT_DONE; default: dev_err(dbc->dev, "Unknown DbC state %d\n", dbc->state); break; diff --git a/drivers/usb/host/xhci-dbgcap.h b/drivers/usb/host/xhci-dbgcap.h index 76170d7a7e7c..2de0dc49a3e9 100644 --- a/drivers/usb/host/xhci-dbgcap.h +++ b/drivers/usb/host/xhci-dbgcap.h @@ -81,7 +81,6 @@ enum dbc_state { DS_ENABLED, DS_CONNECTED, DS_CONFIGURED, - DS_STALLED, }; struct dbc_ep { @@ -89,6 +88,7 @@ struct dbc_ep { struct list_head list_pending; struct xhci_ring *ring; unsigned int direction:1; + unsigned int halted:1; }; #define DBC_QUEUE_SIZE 16 -- 2.25.1

1 month, 3 weeks

3
3
0 0

[PATCH 6.1] gve: Fixes for napi_poll when budget is 0

by Ziwei Xiao

Netpoll will explicitly pass the polling call with a budget of 0 to indicate it's clearing the Tx path only. For the gve_rx_poll and gve_xdp_poll, they were mistakenly taking the 0 budget as the indication to do all the work. Add check to avoid the rx path and xdp path being called when budget is 0. And also avoid napi_complete_done being called when budget is 0 for netpoll. The original fix was merged here: https://lore.kernel.org/r/20231114004144.2022268-1-ziweixiao@google.com Resend it since the original one was not cleanly applied to 6.1 kernel. Fixes: f5cedc84a30d ("gve: Add transmit and receive support") Signed-off-by: Ziwei Xiao <ziweixiao(a)google.com> Reviewed-by: Praveen Kaligineedi <pkaligineedi(a)google.com> Signed-off-by: Praveen Kaligineedi <pkaligineedi(a)google.com> --- drivers/net/ethernet/google/gve/gve_main.c | 7 +++++++ drivers/net/ethernet/google/gve/gve_rx.c | 4 ---- drivers/net/ethernet/google/gve/gve_tx.c | 4 ---- 3 files changed, 7 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/google/gve/gve_main.c b/drivers/net/ethernet/google/gve/gve_main.c index d3f6ad586ba1..8771ccfc69b4 100644 --- a/drivers/net/ethernet/google/gve/gve_main.c +++ b/drivers/net/ethernet/google/gve/gve_main.c @@ -202,6 +202,10 @@ static int gve_napi_poll(struct napi_struct *napi, int budget) if (block->tx) reschedule |= gve_tx_poll(block, budget); + + if (!budget) + return 0; + if (block->rx) { work_done = gve_rx_poll(block, budget); reschedule |= work_done == budget; @@ -242,6 +246,9 @@ static int gve_napi_poll_dqo(struct napi_struct *napi, int budget) if (block->tx) reschedule |= gve_tx_poll_dqo(block, /*do_clean=*/true); + if (!budget) + return 0; + if (block->rx) { work_done = gve_rx_poll_dqo(block, budget); reschedule |= work_done == budget; diff --git a/drivers/net/ethernet/google/gve/gve_rx.c b/drivers/net/ethernet/google/gve/gve_rx.c index 021bbf308d68..639eb6848c7d 100644 --- a/drivers/net/ethernet/google/gve/gve_rx.c +++ b/drivers/net/ethernet/google/gve/gve_rx.c @@ -778,10 +778,6 @@ int gve_rx_poll(struct gve_notify_block *block, int budget) feat = block->napi.dev->features; - /* If budget is 0, do all the work */ - if (budget == 0) - budget = INT_MAX; - if (budget > 0) work_done = gve_clean_rx_done(rx, budget, feat); diff --git a/drivers/net/ethernet/google/gve/gve_tx.c b/drivers/net/ethernet/google/gve/gve_tx.c index 5e11b8236754..bf1ac0d1dc6f 100644 --- a/drivers/net/ethernet/google/gve/gve_tx.c +++ b/drivers/net/ethernet/google/gve/gve_tx.c @@ -725,10 +725,6 @@ bool gve_tx_poll(struct gve_notify_block *block, int budget) u32 nic_done; u32 to_do; - /* If budget is 0, do all the work */ - if (budget == 0) - budget = INT_MAX; - /* In TX path, it may try to clean completed pkts in order to xmit, * to avoid cleaning conflict, use spin_lock(), it yields better * concurrency between xmit/clean than netif's lock. -- 2.47.0.338.g60cca15819-goog

1 month, 3 weeks

3
2
0 0

[PATCH net 14/15] can: mcp251xfd: mcp251xfd_get_tef_len(): work around erratum DS80000789E 6.

by Marc Kleine-Budde

Commit b8e0ddd36ce9 ("can: mcp251xfd: tef: prepare to workaround broken TEF FIFO tail index erratum") introduced mcp251xfd_get_tef_len() to get the number of unhandled transmit events from the Transmit Event FIFO (TEF). As the TEF has no head index, the driver uses the TX-FIFO's tail index instead, assuming that send frames are completed. When calculating the number of unhandled TEF events, that commit didn't take mcp2518fd erratum DS80000789E 6. into account. According to that erratum, the FIFOCI bits of a FIFOSTA register, here the TX-FIFO tail index might be corrupted. However here it seems the bit indicating that the TX-FIFO is empty (MCP251XFD_REG_FIFOSTA_TFERFFIF) is not correct while the TX-FIFO tail index is. Assume that the TX-FIFO is indeed empty if: - Chip's head and tail index are equal (len == 0). - The TX-FIFO is less than half full. (The TX-FIFO empty case has already been checked at the beginning of this function.) - No free buffers in the TX ring. If the TX-FIFO is assumed to be empty, assume that the TEF is full and return the number of elements in the TX-FIFO (which equals the number of TEF elements). If these assumptions are false, the driver might read to many objects from the TEF. mcp251xfd_handle_tefif_one() checks the sequence numbers and will refuse to process old events. Reported-by: Renjaya Raga Zenta <renjaya.zenta(a)formulatrix.com> Closes: https://patch.msgid.link/CAJ7t6HgaeQ3a_OtfszezU=zB-FqiZXqrnATJ3UujNoQJJf7Gg… Fixes: b8e0ddd36ce9 ("can: mcp251xfd: tef: prepare to workaround broken TEF FIFO tail index erratum") Tested-by: Renjaya Raga Zenta <renjaya.zenta(a)formulatrix.com> Cc: stable(a)vger.kernel.org Link: https://patch.msgid.link/20241126-mcp251xfd-fix-length-calculation-v2-1-c2e… Signed-off-by: Marc Kleine-Budde <mkl(a)pengutronix.de> --- drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c | 29 ++++++++++++++++++- 1 file changed, 28 insertions(+), 1 deletion(-) diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c index d3ac865933fd..e94321849fd7 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tef.c @@ -21,6 +21,11 @@ static inline bool mcp251xfd_tx_fifo_sta_empty(u32 fifo_sta) return fifo_sta & MCP251XFD_REG_FIFOSTA_TFERFFIF; } +static inline bool mcp251xfd_tx_fifo_sta_less_than_half_full(u32 fifo_sta) +{ + return fifo_sta & MCP251XFD_REG_FIFOSTA_TFHRFHIF; +} + static inline int mcp251xfd_tef_tail_get_from_chip(const struct mcp251xfd_priv *priv, u8 *tef_tail) @@ -147,7 +152,29 @@ mcp251xfd_get_tef_len(struct mcp251xfd_priv *priv, u8 *len_p) BUILD_BUG_ON(sizeof(tx_ring->obj_num) != sizeof(len)); len = (chip_tx_tail << shift) - (tail << shift); - *len_p = len >> shift; + len >>= shift; + + /* According to mcp2518fd erratum DS80000789E 6. the FIFOCI + * bits of a FIFOSTA register, here the TX-FIFO tail index + * might be corrupted. + * + * However here it seems the bit indicating that the TX-FIFO + * is empty (MCP251XFD_REG_FIFOSTA_TFERFFIF) is not correct + * while the TX-FIFO tail index is. + * + * We assume the TX-FIFO is empty, i.e. all pending CAN frames + * haven been send, if: + * - Chip's head and tail index are equal (len == 0). + * - The TX-FIFO is less than half full. + * (The TX-FIFO empty case has already been checked at the + * beginning of this function.) + * - No free buffers in the TX ring. + */ + if (len == 0 && mcp251xfd_tx_fifo_sta_less_than_half_full(fifo_sta) && + mcp251xfd_get_tx_free(tx_ring) == 0) + len = tx_ring->obj_num; + + *len_p = len; return 0; } -- 2.45.2

1 month, 3 weeks

1
0
0 0

f2fs: fix fiemap failure issue when page size is 16KB

by Daniel Rosenberg

Commit a7a7c1d423a6 ("f2fs: fix fiemap failure issue when page size is 16KB") It resolves an infinite loop in fiemap when using 16k f2fs filesystems. Please apply to stable 6.7-6.12 -Daniel

1 month, 3 weeks

2
1
0 0

[PATCH v4 0/4] media: uvcvideo: Two fixes for async controls

by Ricardo Ribalda

This patchset fixes two bugs with the async controls for the uvc driver. They were found while implementing the granular PM, but I am sending them as a separate patches, so they can be reviewed sooner. They fix real issues in the driver that need to be taken care. Signed-off-by: Ricardo Ribalda <ribalda(a)chromium.org> --- Changes in v4: - Fix implementation of uvc_ctrl_set_handle. - Link to v3: https://lore.kernel.org/r/20241129-uvc-fix-async-v3-0-ab675ce66db7@chromium… Changes in v3: - change again! order of patches. - Introduce uvc_ctrl_set_handle. - Do not change ctrl->handle if it is not NULL. Changes in v2: - Annotate lockdep - ctrl->handle != handle - Change order of patches - Move documentation of mutex - Link to v1: https://lore.kernel.org/r/20241127-uvc-fix-async-v1-0-eb8722531b8c@chromium… --- Ricardo Ribalda (4): media: uvcvideo: Do not replace the handler of an async ctrl media: uvcvideo: Remove dangling pointers media: uvcvideo: Annotate lock requirements for uvc_ctrl_set media: uvcvideo: Remove redundant NULL assignment drivers/media/usb/uvc/uvc_ctrl.c | 62 ++++++++++++++++++++++++++++++++++++---- drivers/media/usb/uvc/uvc_v4l2.c | 2 ++ drivers/media/usb/uvc/uvcvideo.h | 14 +++++++-- 3 files changed, 70 insertions(+), 8 deletions(-) --- base-commit: 72ad4ff638047bbbdf3232178fea4bec1f429319 change-id: 20241127-uvc-fix-async-2c9d40413ad8 Best regards, -- Ricardo Ribalda <ribalda(a)chromium.org>

1 month, 3 weeks

1
4
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror December 2024