What's the status on this issue?  We are rapidly running out of time to get a Mesa update pushed to Chrome.

Fixed on master with:

commit f7aad9da20b13c98f77d6a690b327716f39c0a47
Author: Dave Airlie <airlied@gmail.com>
Date:   Sun Apr 5 16:48:47 2015 +1000

    mesa/teximage: use correct extension for accept stencil texture.
    
    This was using the wrong extension, ARB_stencil_texturing
    doesn't mention any changes in this area.
    
    Fixes "dEQP-GLES3.functional.fbo.completeness.renderable.texture.
    stencil.stencil_index8."
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90751
    Signed-off-by: Dave Airlie <airlied@redhat.com>
    Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
    Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>

I made a mistake, and that patch regressed a ton of tests on Broadwell.

I have a new +2 line patch which should fix those regressions.

Patch to fix the regressions on the mailing list:
http://lists.freedesktop.org/archives/mesa-dev/2015-June/086127.html

Oh, those got fixed a while ago too, by:

commit f83b9e58f6e8a748def367c7d523eb7285b1aeb7
Author: Kenneth Graunke <kenneth@whitecape.org>
Date:   Tue Jun 9 14:33:47 2015 -0700

    i965: Momentarily pretend to support ARB_texture_stencil8 for blits.

So we should be all good here.

Verified and fixed by Mesa 4d35eef326e49cc8da50879d30a1c5088d4775e1 and kernel 4.1.0-rc7_drm-intel-nightly_4643c5_20150613+.
Thanks

This is still reproduceable for us on ChromeOS. 

We added "i965: Momentarily pretend to support ARB_texture_stencil8 for blits." patch but the test "dEQP-GLES3.functional.fbo.completeness.renderable.texture.stencil.stencil_index8" still fails on BDW and BSW.

Mesa used:
c2a0600d5b0645533ba442b5ab879b23c2564a4d: "i965: Don't set NirOptions for stages that will use the vec4 backend."

Plus patches:
- 10.6-i965-do-not-round-line-width-when-multisampling-or-a.patch
- 10.6-mesa-add-GL_RED-GL_RG-support-for-floating-point-tex.patch
- 10.6-i965-Momentarily-pretend-to-support-ARB_texture_sten.patch
- 10.6-Revert-i965-Advertise-a-line-width-of-40.0-on-Cherry.patch

(In reply to Brian Wilson from comment #7)
> This is still reproduceable for us on ChromeOS. 
> 
> We added "i965: Momentarily pretend to support ARB_texture_stencil8 for
> blits." patch but the test
> "dEQP-GLES3.functional.fbo.completeness.renderable.texture.stencil.
> stencil_index8" still fails on BDW and BSW.
> 
> Mesa used:
> c2a0600d5b0645533ba442b5ab879b23c2564a4d: "i965: Don't set NirOptions for
> stages that will use the vec4 backend."
> 
> Plus patches:
> - 10.6-i965-do-not-round-line-width-when-multisampling-or-a.patch
> - 10.6-mesa-add-GL_RED-GL_RG-support-for-floating-point-tex.patch
> - 10.6-i965-Momentarily-pretend-to-support-ARB_texture_sten.patch
> - 10.6-Revert-i965-Advertise-a-line-width-of-40.0-on-Cherry.patch

Do you also have f7aad9d?

Just talked with the integrators and it looks like we might be unclear on the required patches for this.

Is this the only required patch for this?:
f83b9e58f6e8a748def367c7d523eb7285b1aeb7: i965: Momentarily pretend to support ARB_texture_stencil8 for blits.

Or is it: 
f7aad9da20b13c98f77d6a690b327716f39c0a47: mesa/teximage: use correct extension for accept stencil texture.

*and* f83b9e58f6e8a748def367c7d523eb7285b1aeb7

We only integrated f83b9e58f6e8a748def367c7d523eb7285b1aeb7 in our patchset.

Thanks for the clarification.

f7aad9da20b13c98f77d6a690b327716f39c0a47: mesa/teximage: use correct extension
for accept stencil texture.

is the commit that fixes this bug.  It also regressed a ton of tests on Broadwell, due to a mistake I made.

f83b9e58f6e8a748def367c7d523eb7285b1aeb7: i965: Momentarily pretend to support
ARB_texture_stencil8 for blits.

fixes those regressions.

You really want both commits.

Closing as fixed.  Please re-open if you still encounter the problem in released versions.