Add some commands to tpmc.

Add some commands to tpmc.

[Luigi Semenzato, 2010-08-16 21:57:50]

As suggested, I added the GetPermissions command.

[Randall Spangler, 2010-08-16 23:04:47]

http://codereview.chromium.org/3165023/diff/2001/3002
File utility/tpmc.c (right):

http://codereview.chromium.org/3165023/diff/2001/3002#newcode23
utility/tpmc.c:23: int nargs;
As we discussed, add a comment: 
/* Set in main, consumed by handler functions below.  We use global variables so
we can also choose to call Tlcl*() functions directly; they don't take
argv/argc. */

http://codereview.chromium.org/3165023/diff/2001/3002#newcode29
utility/tpmc.c:29: int HexStringToUint32(const char* string, uint32_t* value) {
Why not just use strtoul()?

http://codereview.chromium.org/3165023/diff/2001/3002#newcode51
utility/tpmc.c:51: * overflow.  Returns 0 for success, non-zero for failure.
Might give an example here.  Since you're reversing the bytes below, "0x123456"
seems to set value to {0x56, 0x34, 0x12}, which is a little counter-intuitive.

Could also just pass in "12 34 56", in which case the order of output is more
directly matched, and it's obvious we're not trying to pass in an
endian-specific integer.

http://codereview.chromium.org/3165023/diff/2001/3002#newcode187
utility/tpmc.c:187: for (i = size - 1; i >= 0; i--) {
At least reads and writes are symmetric in that they're starting at the end of
the array.

http://codereview.chromium.org/3165023/diff/2001/3002#newcode230
utility/tpmc.c:230: { "setnvlocked", "setnv", "set the nvLocked flag permanently
(UNDOABLE!)",
UN-UNDOABLE!

http://codereview.chromium.org/3165023/diff/2001/3002#newcode238
utility/tpmc.c:238: { "write", "write", "write to a space [write <index>
<value>]",
write <index> <byte1> [ <byte2> ... ]

[Luigi Semenzato, 2010-08-16 23:59:02]

PTAL

[gauravsh, 2010-08-17 01:49:41]

http://codereview.chromium.org/3165023/diff/7001/3004
File utility/tpmc.c (right):

http://codereview.chromium.org/3165023/diff/7001/3004#newcode35
utility/tpmc.c:35: *value = strtoul(string, &end, 16);
should you be doing a range check here? You shouldn't be assuming that this is a
32-bit system.

http://codereview.chromium.org/3165023/diff/7001/3004#newcode114
utility/tpmc.c:114: uint8_t value[4096];
change this magic value to a symbolic constant

http://codereview.chromium.org/3165023/diff/7001/3004#newcode168
utility/tpmc.c:168: for (i = 0; i < size - 1; i++) {
think I read somewhere in our style guide that we should always use the
pre-increment operator

[Luigi Semenzato, 2010-08-17 16:18:58]

PTAQL!

http://codereview.chromium.org/3165023/diff/7001/3004
File utility/tpmc.c (right):

http://codereview.chromium.org/3165023/diff/7001/3004#newcode35
utility/tpmc.c:35: *value = strtoul(string, &end, 16);
On 2010/08/17 01:49:41, gauravsh wrote:
> should you be doing a range check here? You shouldn't be assuming that this is
a
> 32-bit system.

Do you mean I should try to protect users against the silent truncation of a
number greater than ffffffff, which can already occur?  Then I can't use
strtoul, and I have to parse the arg more carefully.  I wasn't going to bother
for this program.

http://codereview.chromium.org/3165023/diff/7001/3004#newcode168
utility/tpmc.c:168: for (i = 0; i < size - 1; i++) {
On 2010/08/17 01:49:41, gauravsh wrote:
> think I read somewhere in our style guide that we should always use the
> pre-increment operator

From the style guide:

For simple scalar (non-object) values there is no reason to prefer one form and
we allow either. For iterators and other template types, use pre-increment.

[Randall Spangler, 2010-08-17 18:41:20]

LGTM