summaryrefslogtreecommitdiff
path: root/tools/lib/python/kdoc
diff options
context:
space:
mode:
authorDavid Oberhollenzer <david.oberhollenzer@sigma-star.at>2026-07-16 13:39:47 +0200
committerKevin Hilman (TI) <khilman@baylibre.com>2026-08-31 09:54:09 -0700
commit2fabd2f406d0cf787be47b3bbeea99b30004033d (patch)
treec01aaf554ffd1c4417195c6f3bc1a29ec8e2ead3 /tools/lib/python/kdoc
parentcee9395acd8043be0644b25c34bfa86623f2b935 (diff)
downloadlinux-next-2fabd2f406d0cf787be47b3bbeea99b30004033d.tar.gz
linux-next-2fabd2f406d0cf787be47b3bbeea99b30004033d.zip
ARM: dts: ti/omap: dra7: fix PCIe PHY clock divider definition
Commit d0bdd8bb7f35 ("ARM: dts: ti/omap: dra7: fix redundant clock divider definition") removes the clock divider definition from optfclk_pciephy_div, reasoning that the table is redundant, as it contains the values 1 and 2, while also setting ti,max-div to 2 while ti,min-div is implicitly initialized to 1. What that commit missed was the fact that the values in the table are in reverse order, i.e. first 2, then 1. As is clearly described in the documentation, the ordering is signifficant, the actual values for the register setting are derived from the index. If no table is specified, the driver assumes that the register value is divider - 1 (see also `_get_val` in `drivers/clk/ti/divider.c`). What is in fact redundant is the value of ti,max-div that is derived from the table if none is specified. To satisfy device tree semantic linters, it is removed instead. Fixes: d0bdd8bb7f35 ("ARM: dts: ti/omap: dra7: fix redundant clock divider definition") Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at> Link: https://patch.msgid.link/20260716113947.92513-1-david.oberhollenzer@sigma-star.at Signed-off-by: Kevin Hilman (TI) <khilman@baylibre.com>
Diffstat (limited to 'tools/lib/python/kdoc')
0 files changed, 0 insertions, 0 deletions