Dear Chandrashekar, dear Kiran,
Thank you for the patch.
Am 25.05.25 um 07:39 schrieb Kiran K:
From: Chandrashekar Devegowda <chandrashekar.devegowda@xxxxxxxxx>
Modify the driver to post 3 fewer buffers than the maximum rx buffers
(64) allowed for the firmware. This change mitigates a hardware issue
causing a race condition in the firmware, improving stability and data
handling.
Interesting. Please elaborate, which firmware versions are affected, and
if a fix is going to be expected, and how to reproduce the issue, so it
can be tested without and with your patch.
Is the errata published? Why three buffers less and not two or four?
Out of curiosity: Does the Microsoft Windows driver do the same?
Signed-off-by: Chandrashekar Devegowda <chandrashekar.devegowda@xxxxxxxxx>
Signed-off-by: Kiran K <kiran.k@xxxxxxxxx>
---
drivers/bluetooth/btintel_pcie.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/bluetooth/btintel_pcie.c b/drivers/bluetooth/btintel_pcie.c
index 03f13de4a723..14f000e08e1e 100644
--- a/drivers/bluetooth/btintel_pcie.c
+++ b/drivers/bluetooth/btintel_pcie.c
@@ -398,7 +398,10 @@ static int btintel_pcie_start_rx(struct btintel_pcie_data *data)
int i, ret;
struct rxq *rxq = &data->rxq;
- for (i = 0; i < rxq->count; i++) {
+ /* Post (BTINTEL_PCIE_RX_DESCS_COUNT - 3) buffers to overcome the
+ * hardware issues leading to race condition at the firmware.
If you had an errata, it’d be great to document it here to.
I’d remove *the*.
+ */
+ for (i = 0; i < rxq->count - 3; i++) {
ret = btintel_pcie_submit_rx(data);
if (ret)
return ret;
Kind regards,
Paul
