semi合集-English.pdf - 第5509页

SEMI P10-0705 © SEMI 1990, 2005 4 6.23 All <shippable _data> keywor ds can be followed by <ship_to> speci fying the de stination and/ or method of transmission for the transmitted data. Whenever app licable, …

SEMI P10-0705 © SEMI 1990, 2005 3

6.8 Within a given mask set, each mask must have a unique identification (MASK_ID), independent of any titles on

the mask.

6.9 Masks which require multiple writing operations (such as phase shift masks) will find those MASK_ID’s

identified with the same group number by the MULTIWRITE keyword.

6.10 The MASK_ID’s defined under a mask group specify the masks to be built. The MASK_ID’s also identify

which patterns are to appear on the mask. A pattern is placed on the mask if and only if the MASK_ID matches the

LEVEL_ID of a pattern referenced under the mask group’s PLACEMENT_TOP_CELL.

6.11 <cell instance> and <pattern group instance> LOCATION’s position the center of the patterns relative to the

center of the CELL_ID referencing them. The cell referenced by the PLACEMENT_TOP_CELL is positioned at the

center of the mask.

6.12 Job file data may be supplied in place of cell and pattern group data. In this case, the MASK_ID is understood

to correspond to JOB_LEVEL in JOB_NAME.

6.13 For coordinate locations, unless otherwise specified for a given keyword, mask options are relative to the

nominal center of the mask, while cell or pattern options are relative to the center of the cell or pattern. Mask

coordinates are after scaling and mirroring, whereas cell and pattern options are before scaling and mirroring. The

coordinates and directions refer to right-handed, rectangular (Cartesian) coordinates, applied to the substrate being

written. All of these features and the orientation of all other patterns written on the substrate assume that it is from

the perspective of the mask writing tool (i.e., “chrome side up”). The nominal center of the mask is determined using

the bottom and left edges (chrome side up) and assumes nominal substrate dimensions. (See SEMI P1.)

6.14 All data in the structure will be in ASCII. The keyword in each record must include only upper case alphabetic

characters and underscore.

6.15 Records (keyword, data value, comment and new line) may not exceed 256 characters. Some records may be

repeated (e.g., BUSINESS_ADDRESS) in order to allow for longer field requirements; these are explicitly indicated

in the syntax specification.

6.16 Comments may be included in records by preceding them with an exclamation point (!). Any data following an

exclamation point and preceding a carriage return and/or linefeed will be ignored. Such comments are included in

the 256 character record limit. If a record starts with an exclamation point, the entire record will be ignored. Only

printable ASCII characters (plus spaces and tabs) are permitted in comments. Comments are for development and

diagnostic purposes only. In commercial use, comments may always be ignored by the recipient without

consequence. On the other hand, ADDITIONAL_..._INFO records must be evaluated by the recipient and applied to

the masks being ordered, and may justify delay in mask delivery.

6.17 Spaces or tabs preceding the keyword will be ignored. At least one space or tab must precede the data field.

Spaces and tabs imbedded in alphanumeric data fields will be preserved. Spaces or tabs may not be imbedded within

numeric data values, but may be used to separate data values in multi-value data fields. At least one comma must

appear between adjacent data values in multi-value data fields; more than one comma may not appear between

adjacent data values. Spaces or tabs trailing the last data value in a data field will be ignored.

6.18 All numeric record values which require units will be in metric unless otherwise specified in §8. Units of

length or position will be in microns unless otherwise specified.

6.19 Dates will be in the numeric format YYYY-MM-DD. Times will be in the format HH:MM:SSZ or HH:MMZ.

Within the transmitted file, times and dates will always be Coordinated Universal Time (UTC) and times will be

followed by the suffix Z.

6.20 Sizing to achieve a CD_TARGET from a CD_DATA must be applied to all pattern files hierarchically

affected by the CD_TARGET record.

6.21 Binary data manipulation operations expect operands (input pattern files) to align center-to-center. This

requires that DATA_PATTERN_WINDOW be defined for patterns which do not contain an explicit record of file

extents.

6.22 All references to horizontal (or “x”) and vertical (or “y”) correspond to normal Cartesian coordinates and

assume the rotational orientation of the substrate when the mask was written.

SEMI P10-0705 © SEMI 1990, 2005 4

6.23 All <shippable_data> keywords can be followed by <ship_to> specifying the destination and/or method of

transmission for the transmitted data. Whenever applicable, the SEMI <mask_results> file (see §7.5) should be sent

accordingly if electronic transmission is specified.

6.24 <shippable_data> keywords which require reference to specific mask feature locations will reference

CD_SITE_ID and/or REGISTR_MARK_ID which must be unique for each location within all transmissions of

<mask_order> for a given MASK_SET_ID.

6.25 When modifications to this standard are required between ballot cycles (e.g. new data field values “on

request”), users should contact the N.A. SEMI P10 task force leader directly. After a poll of the active task force

members, a tentative recommendation will be posted on the SEMI P10 task force web site at SEMI so that all users

can adopt a common interim solution until a formal ballot can officially resolve the issue.

6.26 Every effort should be made to construct a SEMI P10 file without inconsistencies or ambiguities. If a vendor

detects such flaws in a file from a customer, the vendor should contact the customer to correct the inconsistencies

and/or remove the ambiguities before proceeding with making the mask.

7 Syntax Specification

The following specifies the syntax for the mask order structures. The record order is specific; records must appear in

the sequence shown. Terminology for the individual records and their allowed data values are specified in §8.

7.1 Syntax Symbol Definition — The lower case syntactic names within < > are used to refer to collections of

records. The use of { } around records or collections of records means an arbitrary number of repetitions of the

records, including none at all. The use of [ ] around a record or collection of records means that the collection is

optional.

7.2 Record Syntax  A keyword is followed by a data field which is followed by an optional comment. The record

is terminated by a carriage return and/or linefeed. Each of these elements may be preceded by any combination of

spaces or tabs, which will be ignored. At least one space or tab is required preceding the first data value in the

record. An exclamation point (!) is required to initiate a comment. Exclamation points, carriage returns, and

linefeeds are not allowed in keywords, data values or comments. Note that a record consisting of a comment only is

also allowed. The syntax for individual records is defined as:

<separator> {<separator>} <data_field>]

[{<separator>} ! <comment>]

{<separator>} <new_line>

where “separator” is a space or a tab, “comment” is any string of printable ASCII characters (plus spaces and tabs),

and “new_line” is a carriage return, linefeed or both.

7.3 Multi-value Data Field Syntax — When data_field contains more than one data_value, such as (x,y) pairs or

(x1,y1,x2,y2) double pairs (see §8), the individual values must be separated by a comma and any combination of

spaces or tabs. One comma is required between each data_value, but not more than one comma. Two commas

without an intervening data value are not permissible.

(x,y) = data_value {separator} comma {separator} data_value

(x,y,z) = (x,y) {separator} comma {separator} (z)

(x1,y1,x2,y2) = (x1,y1) {separator} comma {separator} (x2,y2)

(x,y,x1,y1,x2,y2) = (x,y) {separator} comma {separator} (x1,y1,x2,y2)

For data manipulation operations only, a semicolon separates layers from its datatype, and a dash indicates a range

of layers or a range of datatypes.

layer1;datatype6 , layer3-layer4;datatype1-datatype9

7.4 Mask Order Structure Syntax — The following syntax definition uses the upper case keyword to represent the

permissible applications of the corresponding data record within the mask order structure.

<mask_order> = START_ORDER SEMI_REVISION CUSTOMER VENDOR

FILE_DATE_TIME [OPERATOR_NAME]

mask_set> END_ORDER CHECKSUM

<mask_set> = MASK_SET_ID <

mask_set_options> END_MASK_SET_OPTIONS

mask_group>} END_MASK_SET

<mask_set_options> = [BUSINESS_CONTACT] {BUSINESS_ADDRESS} {BUSINESS_PHONE}

{BUSINESS_FAX} [BUSINESS_EMAIL]

[BILLING_CONTACT] {BILLING_ADDRESS}

{BILLING_PHONE} {BILLING_FAX} [BILLING_EMAIL]

[DESIGN_RULE] [FAB_TECHNOLOGY] [WAFER_FAB]

[CUSTOMER_SPEC CUSTOMER_SPEC_REVISION]

{WAIVER} {WAIVER_OVERRIDE}

[PRICE_UNITS]

[SECURITY_CLASS]

[QS9000]

[ENGINEERING_CONTACT] {ENGINEERING ADDRESS} {ENGINEERING_PHONE}

{ENGINEERING_FAX} [ENGINEERING_EMAIL]

[SHIPPING_CONTACT] {SHIPPING_ADDRESS} {SHIPPING_PHONE}

{SHIPPING_FAX} [SHIPPING_EMAIL]

{SHIPPING_METHOD}

[REPAIRS_AUTHORIZED]

shippable_data>]

data>}

standard_pattern>}

[MASK_SET_NAME [MASK_SET_VERSION] ] [ORDER_ID]

{ADDITIONAL_MASK_SET_INFO}

<data> = DATA_MEDIUM [DATA_LOCATION] [DATA_ID]

[DATA_FORMAT [DXF_ANGLE DXF_UNIT] ]

[DATA_DENSITY] [DATA_FILE_SIZE]

[DATA_COMPRESSION] [DATA_CONSOLIDATION] [DATA_ENCRYPTION]

[DATA_CHECKSUM DATA_CHECKSUM_TYPE ]

{<pattern_data>} {<job_data>}

[

FRACTURE_FILE FRACTURE_FILE_FORMAT]

{MEASURE_FILE_NAME [MEASURE_FILE_FORMAT] }

{CD_MATRIX_FILE_NAME [CD_MATRIX_FILE_FORMAT] }

END_DATA_MEDIUM

<pattern_data> = DATA_PATTERN_NAME {PATTERN_FORMAT} [DATA_FILE_NUMBER]

[PATTERN_UNITS] [PATTERN_CHARACTER_SET]