From 3b5510dce1e02e174c3ca831e16fe3f5c9210029 Mon Sep 17 00:00:00 2001
From: Robert Fancsik <frobert@inf.u-szeged.hu>
Date: Fri, 20 Apr 2018 19:50:32 +0200
Subject: [PATCH] Fix SDIO send_command

This patch fixes the problem, while receiving R1b type of response from the card.
In case of R1b response the continuous stream that the card sends must be checked whether the signal indicates busy status and wait until a non zero response.
---
 extras/sdio/sdio.c | 23 +++++++++++++++++++----
 1 file changed, 19 insertions(+), 4 deletions(-)

diff --git a/extras/sdio/sdio.c b/extras/sdio/sdio.c
index 1f273fa..a721442 100644
--- a/extras/sdio/sdio.c
+++ b/extras/sdio/sdio.c
@@ -182,10 +182,6 @@ static uint8_t command(sdio_card_t *card, uint8_t cmd, uint32_t arg)
     wait();
     spi_transfer(BUS, buf, NULL, 6, SPI_8BIT);
 
-    // R1b response
-    if (cmd == CMD12 || cmd == CMD28 || cmd == CMD29)
-        spi_read_byte();
-
     uint8_t res;
     for (uint8_t i = 0; i < MAX_ERR_COUNT; i ++)
     {
@@ -193,6 +189,25 @@ static uint8_t command(sdio_card_t *card, uint8_t cmd, uint32_t arg)
         if (!(res & BV(R1_BUSY)))
             break;
     }
+
+    /** If the response is a "busy" type (R1B), then there’s some
+     * special handling that needs to be done. The card will
+     * output a continuous stream of zeros, so the end of the BUSY
+     * state is signaled by any nonzero response.
+     */
+    if (cmd == CMD12 || cmd == CMD28 || cmd == CMD29)
+    {
+      for (uint8_t i = 0; i < MAX_ERR_COUNT; i ++)
+      {
+          res = spi_read_byte();
+          if (res != 0)
+          {
+            spi_transfer_8(BUS, 0xFF);
+            return SDIO_ERR_NONE;
+          }
+      }
+    }
+    
     return res;
 }