flash/nor/rp2040: Fix incorrect erase bounds calculation

when erase region does not start at 0

Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Signed-off-by: Luke Wren <luke@raspberrypi.com>
Change-Id: I2b9db61e8ac837b6c6431aacf3b73ed3a1772fbc
Reviewed-on: https://review.openocd.org/c/openocd/+/8445
Tested-by: jenkins
Reviewed-by: Jonathan Bell <jonathan@raspberrypi.com>

diff --git a/src/flash/nor/rp2040.c b/src/flash/nor/rp2040.c
index 9151b687898912ce3b477307a5ef47d2e953d855..ed8341d1ff963b53af7e034c0223c360165f11ed 100644 (file)
--- a/src/flash/nor/rp2040.c
+++ b/src/flash/nor/rp2040.c
@@ -821,7 +821,7 @@ static int rp2040_flash_erase(struct flash_bank *bank, unsigned int first, unsig
                goto cleanup_and_return;
 
        uint32_t offset_next = bank->sectors[first].offset;
-       uint32_t offset_last = bank->sectors[last].offset + bank->sectors[last].size - bank->sectors[first].offset;
+       uint32_t offset_last = bank->sectors[last].offset + bank->sectors[last].size;
 
        /* Break erase into multiple calls to avoid timeout on large erase. Choose 128k chunk which has
           fairly low ROM call overhead and empirically seems to avoid the default keep_alive() limit