[pinmux scripts PATCH 1/4] Board CSV import: Support all Tegra124 OD pins

[pinmux scripts PATCH 1/4] Board CSV import: Support all Tegra124 OD pins

[Stephen Warren]

From: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>

Complete the list of OD pins in the CSV -> *.board import script for
Tegra124.

Signed-off-by: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
---
 csv-to-board-tegra124-xlsx.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/csv-to-board-tegra124-xlsx.py b/csv-to-board-tegra124-xlsx.py
index be985c2b83c6..c150fface9d2 100755
--- a/csv-to-board-tegra124-xlsx.py
+++ b/csv-to-board-tegra124-xlsx.py
@@ -219,9 +219,8 @@ with open(csvfile, newline='') as fh:
             print('ERROR: %s: MUX CSV %s not in SOC F0..3 %s' % (ball_name, mux, repr(gpio_pin.funcs)), file=sys.stderr)
             sys.exit(1)
 
-        if ball_name.startswith('ddc_'):
+        if ball_name in ('reset_out_n', 'owr', 'hdmi_int', 'ddc_scl', 'ddc_sda'):
             # These balls' pad type is always OD, so we don't need to set it
-            # FIXME: There are a few other OD type pads to check for
             # FIXME: The SoC data structure should tell us the pad type instead of hard-coding it
             od = False
 
-- 
1.9.1

[pinmux scripts PATCH 2/4] Board CSV import: Put full CSV filename in data structure

[Stephen Warren]

From: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>

Putting the whole filename in to the data structure allows us to keep
different CSV files in different directories. This will allow explicit
separation of public and private CSV files.

The per-board data is also modified to be a dictionary so that additional
fields can be stored in the future.

Signed-off-by: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
---
 csv-to-board-tegra124-xlsx.py | 21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)

diff --git a/csv-to-board-tegra124-xlsx.py b/csv-to-board-tegra124-xlsx.py
index c150fface9d2..d5a8f2f3f645 100755
--- a/csv-to-board-tegra124-xlsx.py
+++ b/csv-to-board-tegra124-xlsx.py
@@ -42,15 +42,26 @@ if args.debug:
 if dbg: print(args)
 
 supported_boards = {
-    'jetson-tk1': 'T124_customer_pinmux_PM375_30Apr2014_v2.csv', # worksheet PM375Beaver_Configuration
-    'norrin': 'PM370_T124_customer_pinmux_1.1.csv', # worksheet Customer_Configuration
-    'venice2': 'Venice2_T124_customer_pinmux_based_on_P4_rev47_2013-07-12.csv', # worksheet Customer_Configuration
+    'jetson-tk1': {
+        # T124_customer_pinmux_PM375_30Apr2014_v2.xlsm worksheet PM375Beaver_Configuration
+        # T124_customer_pinmux.xlsm worksheet Jetson TK1 Configuration
+        # Jetson_TK1_customer_pinmux_release.xlsm worksheet Jetson TK1 Configuration
+        'filename': 'csv/jetson-tk1.csv', # worksheet PM375Beaver_Configuration
+    },
+    'norrin': {
+        # PM370_T124_customer_pinmux_1.1.xlsm worksheet Customer_Configuration
+        'filename': 'nv-internal-data/PM370_T124_customer_pinmux_1.1.csv',
+    },
+    'venice2': {
+        # Venice2_T124_customer_pinmux_based_on_P4_rev47_2013-07-12.xlsm worksheet Customer_Configuration
+        'filename': 'nv-internal-data/Venice2_T124_customer_pinmux_based_on_P4_rev47_2013-07-12.csv',
+    },
 }
 
 if not args.board in supported_boards:
     print('ERROR: Unsupported board %s' % args.board, file=sys.stderr)
     sys.exit(1)
-csvfile = os.path.join('nv-internal-data', supported_boards[args.board])
+board_conf = supported_boards[args.board]
 
 soc = tegra_pmx_soc_parser.load_soc('tegra124')
 
@@ -136,7 +147,7 @@ def rcv_sel_munge(d):
 
 found_header = False
 pin_table = []
-with open(csvfile, newline='') as fh:
+with open(board_conf['filename'], newline='') as fh:
     csv = csv.reader(fh)
     lnum = 0
     for row in csv:
-- 
1.9.1

[pinmux scripts PATCH 3/4] Board CSV import: Support either 0- or 1-based RSVD numbers

[Stephen Warren]

From: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>

The public Jetson TK1 pinmux spreadsheet will use 1-based RSVD numbers
whereas the other internal board spreadsheets aren't (currently?)
updated, and hence will continue to use 0-based RSVD numbering. Support
either numbering scheme in the import script.

Signed-off-by: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
---
 csv-to-board-tegra124-xlsx.py | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/csv-to-board-tegra124-xlsx.py b/csv-to-board-tegra124-xlsx.py
index d5a8f2f3f645..a4d414c3e991 100755
--- a/csv-to-board-tegra124-xlsx.py
+++ b/csv-to-board-tegra124-xlsx.py
@@ -47,6 +47,7 @@ supported_boards = {
         # T124_customer_pinmux.xlsm worksheet Jetson TK1 Configuration
         # Jetson_TK1_customer_pinmux_release.xlsm worksheet Jetson TK1 Configuration
         'filename': 'csv/jetson-tk1.csv', # worksheet PM375Beaver_Configuration
+        'rsvd_0based': False,
     },
     'norrin': {
         # PM370_T124_customer_pinmux_1.1.xlsm worksheet Customer_Configuration
@@ -62,6 +63,9 @@ if not args.board in supported_boards:
     print('ERROR: Unsupported board %s' % args.board, file=sys.stderr)
     sys.exit(1)
 board_conf = supported_boards[args.board]
+if not 'rsvd_0based' in board_conf:
+    # FIXME: This should default to False for some future chip
+    board_conf['rsvd_0based'] = True
 
 soc = tegra_pmx_soc_parser.load_soc('tegra124')
 
@@ -108,7 +112,9 @@ def func_munge(f):
         return 'sdmmc2'
     if f in ('ir3_rxd', 'ir3_txd'):
         return 'irda'
-    return rsvd_0base_to_1base(f)
+    if board_conf['rsvd_0based']:
+        return rsvd_0base_to_1base(f)
+    return f
 
 def pupd_munge(d):
     return {
-- 
1.9.1

[pinmux scripts PATCH 4/4] Board CSV Import: Allow board configuration overrides on cmdline

[Stephen Warren]

From: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>

Add --csv FILENAME, --rsvd-0based, and --rsvd-1based command-line options
to the board CSV import script. This is especially useful for Jetson TK1,
since different spreadsheets use 0- and 1-based RSVD numbering, so the user
may require an option to easily specify which to use.

Signed-off-by: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
---
 csv-to-board-tegra124-xlsx.py | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/csv-to-board-tegra124-xlsx.py b/csv-to-board-tegra124-xlsx.py
index a4d414c3e991..eb381656e40d 100755
--- a/csv-to-board-tegra124-xlsx.py
+++ b/csv-to-board-tegra124-xlsx.py
@@ -35,6 +35,9 @@ dbg = False
 parser = argparse.ArgumentParser(description='Create a board config' +
     'from a CSV version of the Venice2 pinmux spreadsheet')
 parser.add_argument('--debug', action='store_true', help='Turn on debugging prints')
+parser.add_argument('--csv', default=argparse.SUPPRESS, help='CSV file to parse')
+parser.add_argument('--rsvd-0based', action='store_true', dest='rsvd_0based', default=argparse.SUPPRESS, help='Assume 0-based RSVD numbering')
+parser.add_argument('--rsvd-1based', action='store_false', dest='rsvd_0based', default=argparse.SUPPRESS, help='Assume 1-based RSVD numbering')
 parser.add_argument('board', help='Board name')
 args = parser.parse_args()
 if args.debug:
@@ -66,6 +69,11 @@ board_conf = supported_boards[args.board]
 if not 'rsvd_0based' in board_conf:
     # FIXME: This should default to False for some future chip
     board_conf['rsvd_0based'] = True
+if 'csv' in args:
+    board_conf['filename'] = args.csv
+if 'rsvd_0based' in args:
+    board_conf['rsvd_0based'] = args.rsvd_0based
+if dbg: print(board_conf)
 
 soc = tegra_pmx_soc_parser.load_soc('tegra124')
 
-- 
1.9.1

Re: [pinmux scripts PATCH 2/4] Board CSV import: Put full CSV filename in data structure

[Stephen Warren]

On 10/14/2014 11:11 AM, Stephen Warren wrote:
> From: Stephen Warren <swarren-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org>
>
> Putting the whole filename in to the data structure allows us to keep
> different CSV files in different directories. This will allow explicit
> separation of public and private CSV files.
>
> The per-board data is also modified to be a dictionary so that additional
> fields can be stored in the future.

I have applied this series.

>   supported_boards = {
> -    'jetson-tk1': 'T124_customer_pinmux_PM375_30Apr2014_v2.csv', # worksheet PM375Beaver_Configuration
> -    'norrin': 'PM370_T124_customer_pinmux_1.1.csv', # worksheet Customer_Configuration
> -    'venice2': 'Venice2_T124_customer_pinmux_based_on_P4_rev47_2013-07-12.csv', # worksheet Customer_Configuration
> +    'jetson-tk1': {
> +        # T124_customer_pinmux_PM375_30Apr2014_v2.xlsm worksheet PM375Beaver_Configuration
> +        # T124_customer_pinmux.xlsm worksheet Jetson TK1 Configuration
> +        # Jetson_TK1_customer_pinmux_release.xlsm worksheet Jetson TK1 Configuration
> +        'filename': 'csv/jetson-tk1.csv', # worksheet PM375Beaver_Configuration

I removed that stale comment on the filename: line, and added a note at 
the end of all the comments in this data structure re: which files use 
0- vs. 1-based rsvd numbering.