403579 - "ASSERTION: Attempted to subtract [n - nscoord_MAX]" with percentage margin in table cell

Status: RESOLVED FIXED

(Core :: Layout: Tables, defect)

Description

[Jesse Ruderman]

Attachment: testcase

Loading the testcase triggers four assertions:

###!!! ASSERTION: Doing nscoord addition with values > nscoord_MAX: 'a < nscoord_MAX && b < nscoord_MAX', file ../../dist/include/gfx/nsCoord.h, line 151

###!!! ASSERTION: nscoord addition will reach or pass nscoord_MAX: '(PRInt64)a + (PRInt64)b < (PRInt64)nscoord_MAX', file ../../dist/include/gfx/nsCoord.h, line 153

###!!! ASSERTION: Doing nscoord subtraction with values > nscoord_MAX: 'a < nscoord_MAX && b < nscoord_MAX', file ../../dist/include/gfx/nsCoord.h, line 208

###!!! ASSERTION: Attempted to subtract [n - nscoord_MAX]: 'Not Reached', file ../../dist/include/gfx/nsCoord.h, line 194

Comment 1

[Daniel Holbert [:dholbert]]

Attachment: fix

Comment 2

[Daniel Holbert [:dholbert]]

Comment on attachment 288556 [details] [diff] [review]
fix

First chunk is similar to patch for bug 402893, but for the collective nonpct columns.

Second chunk is necessary because guess_pref can be nscoord_MAX if we hit this line (at the end of the context region) in an earlier iteration of the loop:

>             guess_pref = NSCoordSaturatingAdd(guess_pref, pref_width);

Comment 3

[Mike Schroepfer]

Comment on attachment 288556 [details] [diff] [review]
fix

dholbert out of p1-2 bugs? :-)

Comment 4

[Daniel Holbert [:dholbert]]

(In reply to comment #3)
> dholbert out of p1-2 bugs? :-)

Was just fixing these as a mental break, 'cause they're pretty quick fixes and are in code I'm very familiar with. :)

Patch checked in:
BasicTableLayoutStrategy.cpp
new revision: 3.263; previous revision: 3.262

Comment 5

[Jesse Ruderman]

Crashtest checked in.