Re: [PATCH net v4 2/3] net: ethernet: oa_tc6: Improvements to error recovery
From: Parthiban Veerasooran <parthiban.veerasooran@microchip.com>
Date: 2026-07-22 10:09:17
Also in:
lkml
Imperative mood for the subject: "Improve error recovery" (netdev convention). On 21/07/26 7:50 am, Selvamani Rajagopal via B4 Relay wrote:
quoted hunk ↗ jump to hunk
static int oa_tc6_prcs_rx_chunk_payload(struct oa_tc6 *tc6, u8 *data,@@ -931,8 +966,7 @@ static int oa_tc6_prcs_rx_chunk_payload(struct oa_tc6 *tc6, u8 *data, /* Process the chunk with only rx frame end */ if (end_valid && !start_valid) { size = end_byte_offset + 1; - oa_tc6_prcs_rx_frame_end(tc6, data, size); - return 0; + return oa_tc6_prcs_rx_frame_end(tc6, data, size); } /* Process the chunk with previous rx frame end and next rx frame@@ -946,6 +980,14 @@ static int oa_tc6_prcs_rx_chunk_payload(struct oa_tc6 *tc6, u8 *data, if (tc6->rx_skb) { size = end_byte_offset + 1; oa_tc6_prcs_rx_frame_end(tc6, data, size);
oa_tc6_prcs_rx_frame_end() now returns int, but this drops the return — every other call in this patch propagates it. Is it deliberate?
quoted hunk ↗ jump to hunk
+ + /* Purpose of rx_buf_overflow is make the + * code to look for new frame. At this + * stage, we have a new frame to process. + * So, making it false, in case it is set + * to true by oa_tc6_prcs_rx_frame_end. + */ + tc6->rx_buf_overflow = false; } size = OA_TC6_CHUNK_PAYLOAD_SIZE - start_byte_offset; return oa_tc6_prcs_rx_frame_start(tc6,@@ -954,9 +996,7 @@ static int oa_tc6_prcs_rx_chunk_payload(struct oa_tc6 *tc6, u8 *data, } /* Process the chunk with ongoing rx frame data */ - oa_tc6_prcs_ongoing_rx_frame(tc6, data, footer); - - return 0; + return oa_tc6_prcs_ongoing_rx_frame(tc6, data, footer);
footer is unused here. Either drop the unused parameter or a note why it stays. Best regards, Parthiban V
}