[Bug 111600] [CI][RESUME] igt@* - dmesg-warn / dmesg-fail - *ERROR* CPU pipe [ABC] FIFO underrun

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Mon Sep 16 06:54:40 UTC 2019


https://bugs.freedesktop.org/show_bug.cgi?id=111600

--- Comment #4 from Stanislav Lisovskiy <stanislav.lisovskiy at intel.com> ---
I have looked and there seems to be no memory bandwidth check working for tgl,
currently intel_max_data_rate function which is used for final comparation to
accept/reject display configuration looks like this(with current drm-tip):

        if (IS_GEN(dev_priv, 11))
                /*
                 * FIXME with SAGV disabled maybe we can assume
                 * point 1 will always be used? Seems to match
                 * the behaviour observed in the wild.
                 */
                return min3(icl_max_bw(dev_priv, num_planes, 0),
                            icl_max_bw(dev_priv, num_planes, 1),
                            icl_max_bw(dev_priv, num_planes, 2));
        else
                return UINT_MAX;
There are other similar parts in intel_bw.c also. Once I modified those to
query pcode and calculate bandwidth same way as for ICL, FIFO underruns
disappeared. 

So my guess is that we are simply lacking bandwidth part here. Do we have any
plans on enabling that bandwidth code for TGL?

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20190916/4bbf354f/attachment.html>


More information about the intel-gfx-bugs mailing list