diff options
| author | Luca Coelho <luciano.coelho@intel.com> | 2026-06-09 12:52:39 +0300 |
|---|---|---|
| committer | Luca Coelho <luciano.coelho@intel.com> | 2026-08-13 11:20:22 +0300 |
| commit | 5655c87693f6a6006208197a4e3cb29f1253087f (patch) | |
| tree | 1279baa0374c01cd3ccc53b6f610deee4fb4d9c7 /tools/perf/scripts/python | |
| parent | 279b0df165e2ec4f853de03b3ee1c7961ac23513 (diff) | |
| download | linux-5655c87693f6a6006208197a4e3cb29f1253087f.tar.gz linux-5655c87693f6a6006208197a4e3cb29f1253087f.zip | |
drm/i915/display: remove unnecessary PHY_NONE definition
PHY_NONE is not really used, but we define it and, thus, need to check
for it in a few places we use phy. The only potential places where
phy may become PHY_NONE, is in intel_port_to_phy(), where it derives
from port, which can be PORT_NONE. Many of its callers don't check
for PHY_NONE, which can cause unknown behavior. Additionally, this
can only happen if the encoder used has PORT_NONE, which should not be
the case either, without unexpected consequences.
Remove the PHY_NONE definition entirely and add a couple of WARNs at
the relevant places, just to be sure.
Reviewed-by: Michał Grzelak <michal.grzelak@intel.com>
Link: https://patch.msgid.link/20260609095525.570614-5-luciano.coelho@intel.com
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Diffstat (limited to 'tools/perf/scripts/python')
0 files changed, 0 insertions, 0 deletions
