FAILED: patch "[PATCH] xhci: dbc: Fix full DbC transfer ring after several" failed to apply to 5.10-stable tree
The patch below does not apply to the 5.10-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@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-5.10.y git checkout FETCH_HEAD git cherry-pick -x a5c98e8b1398534ae1feb6e95e2d3ee5215538ed # <resolve conflicts, build, test, etc.> git commit -s git send-email --to 'stable@vger.kernel.org' --in-reply-to '2025091731-subtotal-outcome-6092@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^..
Possible dependencies:
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From a5c98e8b1398534ae1feb6e95e2d3ee5215538ed Mon Sep 17 00:00:00 2001 From: Mathias Nyman mathias.nyman@linux.intel.com Date: Tue, 2 Sep 2025 13:53:05 +0300 Subject: [PATCH] xhci: dbc: Fix full DbC transfer ring after several reconnects
Pending requests will be flushed on disconnect, and the corresponding TRBs will be turned into No-op TRBs, which are ignored by the xHC controller once it starts processing the ring.
If the USB debug cable repeatedly disconnects before ring is started then the ring will eventually be filled with No-op TRBs. No new transfers can be queued when the ring is full, and driver will print the following error message:
"xhci_hcd 0000:00:14.0: failed to queue trbs"
This is a normal case for 'in' transfers where TRBs are always enqueued in advance, ready to take on incoming data. If no data arrives, and device is disconnected, then ring dequeue will remain at beginning of the ring while enqueue points to first free TRB after last cancelled No-op TRB. s Solve this by reinitializing the rings when the debug cable disconnects and DbC is leaving the configured state. Clear the whole ring buffer and set enqueue and dequeue to the beginning of ring, and set cycle bit to its initial state.
Cc: stable@vger.kernel.org Fixes: dfba2174dc42 ("usb: xhci: Add DbC support in xHCI driver") Signed-off-by: Mathias Nyman mathias.nyman@linux.intel.com Link: https://lore.kernel.org/r/20250902105306.877476-3-mathias.nyman@linux.intel.... Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
diff --git a/drivers/usb/host/xhci-dbgcap.c b/drivers/usb/host/xhci-dbgcap.c index d0faff233e3e..63edf2d8f245 100644 --- a/drivers/usb/host/xhci-dbgcap.c +++ b/drivers/usb/host/xhci-dbgcap.c @@ -462,6 +462,25 @@ static void xhci_dbc_ring_init(struct xhci_ring *ring) xhci_initialize_ring_info(ring); }
+static int xhci_dbc_reinit_ep_rings(struct xhci_dbc *dbc) +{ + struct xhci_ring *in_ring = dbc->eps[BULK_IN].ring; + struct xhci_ring *out_ring = dbc->eps[BULK_OUT].ring; + + if (!in_ring || !out_ring || !dbc->ctx) { + dev_warn(dbc->dev, "Can't re-init unallocated endpoints\n"); + return -ENODEV; + } + + xhci_dbc_ring_init(in_ring); + xhci_dbc_ring_init(out_ring); + + /* set ep context enqueue, dequeue, and cycle to initial values */ + xhci_dbc_init_ep_contexts(dbc); + + return 0; +} + static struct xhci_ring * xhci_dbc_ring_alloc(struct device *dev, enum xhci_ring_type type, gfp_t flags) { @@ -885,7 +904,7 @@ static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) dev_info(dbc->dev, "DbC cable unplugged\n"); dbc->state = DS_ENABLED; xhci_dbc_flush_requests(dbc); - + xhci_dbc_reinit_ep_rings(dbc); return EVT_DISC; }
@@ -895,7 +914,7 @@ static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) writel(portsc, &dbc->regs->portsc); dbc->state = DS_ENABLED; xhci_dbc_flush_requests(dbc); - + xhci_dbc_reinit_ep_rings(dbc); return EVT_DISC; }
From: Mathias Nyman mathias.nyman@linux.intel.com
[ Upstream commit 220a0ffde02f962c13bc752b01aa570b8c65a37b ]
Decouple allocation of endpoint ring buffer from initialization of the buffer, and initialization of endpoint context parts from from the rest of the contexts.
It allows driver to clear up and reinitialize endpoint rings after disconnect without reallocating everything.
This is a prerequisite for the next patch that prevents the transfer ring from filling up with cancelled (no-op) TRBs if a debug cable is reconnected several times without transferring anything.
Cc: stable@vger.kernel.org Fixes: dfba2174dc42 ("usb: xhci: Add DbC support in xHCI driver") Signed-off-by: Mathias Nyman mathias.nyman@linux.intel.com Link: https://lore.kernel.org/r/20250902105306.877476-2-mathias.nyman@linux.intel.... Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/usb/host/xhci-dbgcap.c | 71 ++++++++++++++++++++++------------ 1 file changed, 46 insertions(+), 25 deletions(-)
diff --git a/drivers/usb/host/xhci-dbgcap.c b/drivers/usb/host/xhci-dbgcap.c index 03f047f5508bf..43106b5840b2d 100644 --- a/drivers/usb/host/xhci-dbgcap.c +++ b/drivers/usb/host/xhci-dbgcap.c @@ -86,13 +86,34 @@ static u32 xhci_dbc_populate_strings(struct dbc_str_descs *strings) return string_length; }
+static void xhci_dbc_init_ep_contexts(struct xhci_dbc *dbc) +{ + struct xhci_ep_ctx *ep_ctx; + unsigned int max_burst; + dma_addr_t deq; + + max_burst = DBC_CTRL_MAXBURST(readl(&dbc->regs->control)); + + /* Populate bulk out endpoint context: */ + ep_ctx = dbc_bulkout_ctx(dbc); + deq = dbc_bulkout_enq(dbc); + ep_ctx->ep_info = 0; + ep_ctx->ep_info2 = dbc_epctx_info2(BULK_OUT_EP, 1024, max_burst); + ep_ctx->deq = cpu_to_le64(deq | dbc->ring_out->cycle_state); + + /* Populate bulk in endpoint context: */ + ep_ctx = dbc_bulkin_ctx(dbc); + deq = dbc_bulkin_enq(dbc); + ep_ctx->ep_info = 0; + ep_ctx->ep_info2 = dbc_epctx_info2(BULK_IN_EP, 1024, max_burst); + ep_ctx->deq = cpu_to_le64(deq | dbc->ring_in->cycle_state); +} + static void xhci_dbc_init_contexts(struct xhci_dbc *dbc, u32 string_length) { struct dbc_info_context *info; - struct xhci_ep_ctx *ep_ctx; u32 dev_info; - dma_addr_t deq, dma; - unsigned int max_burst; + dma_addr_t dma;
if (!dbc) return; @@ -106,20 +127,8 @@ static void xhci_dbc_init_contexts(struct xhci_dbc *dbc, u32 string_length) info->serial = cpu_to_le64(dma + DBC_MAX_STRING_LENGTH * 3); info->length = cpu_to_le32(string_length);
- /* Populate bulk out endpoint context: */ - ep_ctx = dbc_bulkout_ctx(dbc); - max_burst = DBC_CTRL_MAXBURST(readl(&dbc->regs->control)); - deq = dbc_bulkout_enq(dbc); - ep_ctx->ep_info = 0; - ep_ctx->ep_info2 = dbc_epctx_info2(BULK_OUT_EP, 1024, max_burst); - ep_ctx->deq = cpu_to_le64(deq | dbc->ring_out->cycle_state); - - /* Populate bulk in endpoint context: */ - ep_ctx = dbc_bulkin_ctx(dbc); - deq = dbc_bulkin_enq(dbc); - ep_ctx->ep_info = 0; - ep_ctx->ep_info2 = dbc_epctx_info2(BULK_IN_EP, 1024, max_burst); - ep_ctx->deq = cpu_to_le64(deq | dbc->ring_in->cycle_state); + /* Populate bulk in and out endpoint contexts: */ + xhci_dbc_init_ep_contexts(dbc);
/* Set DbC context and info registers: */ lo_hi_writeq(dbc->ctx->dma, &dbc->regs->dccp); @@ -421,6 +430,23 @@ dbc_alloc_ctx(struct device *dev, gfp_t flags) return ctx; }
+static void xhci_dbc_ring_init(struct xhci_ring *ring) +{ + struct xhci_segment *seg = ring->first_seg; + + /* clear all trbs on ring in case of old ring */ + memset(seg->trbs, 0, TRB_SEGMENT_SIZE); + + /* Only event ring does not use link TRB */ + if (ring->type != TYPE_EVENT) { + union xhci_trb *trb = &seg->trbs[TRBS_PER_SEGMENT - 1]; + + trb->link.segment_ptr = cpu_to_le64(ring->first_seg->dma); + trb->link.control = cpu_to_le32(LINK_TOGGLE | TRB_TYPE(TRB_LINK)); + } + xhci_initialize_ring_info(ring, 1); +} + static struct xhci_ring * xhci_dbc_ring_alloc(struct device *dev, enum xhci_ring_type type, gfp_t flags) { @@ -449,15 +475,10 @@ xhci_dbc_ring_alloc(struct device *dev, enum xhci_ring_type type, gfp_t flags)
seg->dma = dma;
- /* Only event ring does not use link TRB */ - if (type != TYPE_EVENT) { - union xhci_trb *trb = &seg->trbs[TRBS_PER_SEGMENT - 1]; - - trb->link.segment_ptr = cpu_to_le64(dma); - trb->link.control = cpu_to_le32(LINK_TOGGLE | TRB_TYPE(TRB_LINK)); - } INIT_LIST_HEAD(&ring->td_list); - xhci_initialize_ring_info(ring, 1); + + xhci_dbc_ring_init(ring); + return ring; dma_fail: kfree(seg);
From: Mathias Nyman mathias.nyman@linux.intel.com
[ Upstream commit a5c98e8b1398534ae1feb6e95e2d3ee5215538ed ]
Pending requests will be flushed on disconnect, and the corresponding TRBs will be turned into No-op TRBs, which are ignored by the xHC controller once it starts processing the ring.
If the USB debug cable repeatedly disconnects before ring is started then the ring will eventually be filled with No-op TRBs. No new transfers can be queued when the ring is full, and driver will print the following error message:
"xhci_hcd 0000:00:14.0: failed to queue trbs"
This is a normal case for 'in' transfers where TRBs are always enqueued in advance, ready to take on incoming data. If no data arrives, and device is disconnected, then ring dequeue will remain at beginning of the ring while enqueue points to first free TRB after last cancelled No-op TRB. s Solve this by reinitializing the rings when the debug cable disconnects and DbC is leaving the configured state. Clear the whole ring buffer and set enqueue and dequeue to the beginning of ring, and set cycle bit to its initial state.
Cc: stable@vger.kernel.org Fixes: dfba2174dc42 ("usb: xhci: Add DbC support in xHCI driver") Signed-off-by: Mathias Nyman mathias.nyman@linux.intel.com Link: https://lore.kernel.org/r/20250902105306.877476-3-mathias.nyman@linux.intel.... Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/usb/host/xhci-dbgcap.c | 23 +++++++++++++++++++++-- 1 file changed, 21 insertions(+), 2 deletions(-)
diff --git a/drivers/usb/host/xhci-dbgcap.c b/drivers/usb/host/xhci-dbgcap.c index 43106b5840b2d..64fdaf468eda1 100644 --- a/drivers/usb/host/xhci-dbgcap.c +++ b/drivers/usb/host/xhci-dbgcap.c @@ -447,6 +447,25 @@ static void xhci_dbc_ring_init(struct xhci_ring *ring) xhci_initialize_ring_info(ring, 1); }
+static int xhci_dbc_reinit_ep_rings(struct xhci_dbc *dbc) +{ + struct xhci_ring *in_ring = dbc->eps[BULK_IN].ring; + struct xhci_ring *out_ring = dbc->eps[BULK_OUT].ring; + + if (!in_ring || !out_ring || !dbc->ctx) { + dev_warn(dbc->dev, "Can't re-init unallocated endpoints\n"); + return -ENODEV; + } + + xhci_dbc_ring_init(in_ring); + xhci_dbc_ring_init(out_ring); + + /* set ep context enqueue, dequeue, and cycle to initial values */ + xhci_dbc_init_ep_contexts(dbc); + + return 0; +} + static struct xhci_ring * xhci_dbc_ring_alloc(struct device *dev, enum xhci_ring_type type, gfp_t flags) { @@ -871,7 +890,7 @@ static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) dev_info(dbc->dev, "DbC cable unplugged\n"); dbc->state = DS_ENABLED; xhci_dbc_flush_requests(dbc); - + xhci_dbc_reinit_ep_rings(dbc); return EVT_DISC; }
@@ -881,7 +900,7 @@ static enum evtreturn xhci_dbc_do_handle_events(struct xhci_dbc *dbc) writel(portsc, &dbc->regs->portsc); dbc->state = DS_ENABLED; xhci_dbc_flush_requests(dbc); - + xhci_dbc_reinit_ep_rings(dbc); return EVT_DISC; }
linux-stable-mirror@lists.linaro.org
-
gregkh＠linuxfoundation.org -
Sasha Levin