Most of the issues are minor at this point. Overall I'm optimistic about this approach!

There is a problem with empty comments in flashchips.c (L1322, for example), however the new version of gerrit seems too slow to be able to comment on that (?!). I'll look into it a bit more so we don't create superfluous newlines.

Patch set 2:Code-Review -1

View Change

8 comments:

• File cbtable.c:

  • Patch Set #2, Line 206: ;

    This semicolon is not needed and should be removed from the original file.

• File hwaccess.h:

  • Patch Set #2, Line 42: #define ___constant_swab8(x) ((uint8_t)( \

    These are a bit funky since the leftmost cast is on the first line. A couple of ways to make them easier for uncrustify to handle:

    • Put the cast on an indented newline
    • Make this into an inline function. IMO this is better since then we can avoid all these casts and have the compiler spit out an error if somehow the wrong type is input.

• File ich_descriptors.h:

  • Patch Set #2, Line 184: : 3,

    uncrustify does this for unnamed fields in bitfields. The easy workaround is to give each a name, e.g. rsvd1, rsvd2, etc.

• File ich_descriptors.c:

  • Patch Set #2, Line 217: static const char *const freq_str[3][8] = { {

    The indent in this case is based on the rightmost brace (for the first array of strings). Moving it to a newline sets it to the left.

• File jlink_spi.c:

  • Patch Set #2, Line 227:

    		if (ret != JAYLINK_OK) {
    			msg_perr("jaylink_parse_serial_number() failed: %s.\n", jaylink_strerror(ret));
    			free(arg);
    			return 1;
    		}
    

    This seems redundant with the preceding if-statement, and should be fixed in another patch.

• File serial.c:

  • Patch Set #2, Line 112: {

    BAUDRATE(0) can be used to make this a single line, if desired.

• File spi25_statusreg.c:

  • Patch Set #2, Line 454: msg_cdbg("no sectors are protected\n");

    This is fixed by putting the case labels in parens, i.e. `case (0x0 << 2):`

• File uncrustify.cfg:

  • Patch Set #2, Line 11: to

    of

To view, visit change 42556. To unsubscribe, or for help writing mail filters, visit settings.