In bug 763689, I'm adding support for the new min-width / min-height: auto keyword. Before that bug's patches, we don't support "auto" for these properties. However, nsHTMLReflowState::InitResizeFlags currently has this check: > 533 bool dependsOnCBHeight = [...] > 537 (mStylePosition->MinHeightDependsOnContainer() && > 538 // FIXME: condition this on not-abspos? > 539 mStylePosition->mMinHeight.GetUnit() != eStyleUnit_Auto) || http://mxr.mozilla.org/mozilla-central/source/layout/generic/nsHTMLReflowState.cpp#533 from this cset: http://hg.mozilla.org/mozilla-central/rev/fcc2aa4bd451#l6.11 That "mMinHeight.GetUnit() != eStyleUnit_Auto" check is trivially true right now. Moreover, after bug 763689 lets us have "auto" in computed-style, we explicitly *won't* want the check for 'auto' here. (since the used value of "min-height:auto" is either 0 or the item's intrinsic height -- it never depends on the container's height). So, I think we want to just drop this line.

(In reply to Daniel Holbert [:dholbert] from comment #0) > That "mMinHeight.GetUnit() != eStyleUnit_Auto" check is trivially true right > now. That's true of the mMaxHeight check in the same chunk of code, too, actually. max-height doesn't accept "auto", as shown here: http://mxr.mozilla.org/mozilla-central/source/layout/style/nsCSSPropList.h#2064

(reposting patch with 8 lines of context, instead of 3, to make it clearer)

Comment on attachment 645904 [details] [diff] [review] fix v1a r=dbaron