Re: [sv-ec] erratum 275 (strings) for section 3.7 updated

Comments:

1. On bottom of page 2, top of page 3:

bit [11:0] b = 12'ha41 ;
string s2 = string' (b) ; // sets s2 to 'h0a41

I believe the 2nd line should be

string s2 = string' (b) ; // sets s2 to 16'h0a41

as 'h0a41 is 32 bits.

2. On bottom of page 3:

r = {"H",""}; // yields "H\0" => "" is converted to 8b0
b = {"H",""}; // yields "H" => "" is the empty string
a[0] = "h"; // OK same as a[0] = "cough" ;

The symbol "=>" is not appropriate here.
That symbol means "implies", which is not the case here.
A simple period would be enough.
In the 3rd line above, there should be a period or a comma between "OK" and
"same as".
Also, in all the examples, not just those quoted here, all quotation marks
should be straight up and down.

3. In Table 3-2 (top of page 4):

"If both operands are string literals, the expression is the same Verilog
equality operator for integer types."
Besides being incorrect (an expression is not an operator), it is not clear what
the intended meaning is.

4. Same place: 'The special value " " is allowed.'

Was this supposed to be the empty string? There is a space here between the two
quotation marks,
so it is not the empty string.

5. Page 4, "Replace Section 3.7.7" should be "Replace Section 3.7.2".

6. 3.7.2, page 4 bottom:

"If the second argument to putc is zero, the string is unaffected: if c is zero,
then str is unchanged."

The second part of the sentence is redundant.

7. Page 5:

The first "Replace Section 3.76" should refer to 3.7.6, the second to 3.7.7.

Shalom

Mehdi Mohtashemi wrote:

> Hi everyone,
> In the champions meeting on Tuesday, 12/21/2004, erratum 275 and
> its related errata, 197, 203, 271 and 334 were discussed. An action
> item was created for sv-ec to:
>
> 1. Update the proposal for 275 so that it is more clear what the
> actual changes are to the LRM.
>
> 2. Update the superceded errata such that it is clear what in them
> got superceded.
>
> I have placed the merged text of all errata related to string
> (section 3.7) in proposal for 275 on the database for members to
> review.
> We will keep erratum 275 with its merged text proposal and
> The next step is to keep errata 275 with its merged text,
> close off all the others (197, 203, 271, 334).
>
> Link to erratum 275:
> http://www.eda.org/svdb/bug_view_page.php?bug_id=0000275
>
> At the next champions meeting, January 4th, 2005, this erratum 275
> will be reviewed.
>
> Happy holidays....
> - Mehdi

--
Shalom Bresticker                        Shalom.Bresticker @freescale.com
Design & Verification Methodology                    Tel: +972 9  9522268
Freescale Semiconductor Israel, Ltd.                 Fax: +972 9  9522890
POB 2208, Herzlia 46120, ISRAEL                     Cell: +972 50 5441478
[ ]Freescale Internal Use Only      [ ]Freescale Confidential Proprietary