Home
last modified time | relevance | path

Searched hist:d7ea335c (Results 1 – 3 of 3) sorted by relevance

/linux-6.15/arch/mips/include/asm/
H A Dtime.hd7ea335c Thu Nov 15 05:34:17 GMT 2012 Steven J. Hill <[email protected]> MIPS: Remove usage of CSRC_R4K_LIB config option.

Manuel Lauss <[email protected]> writes:

I introduced it as a fallback because early revisions of Alchemy hardware
we shipped had a non-functional 32kHz timer and had to rely on the r4k
timer instead. Previously the r4k timer was initialized regardless, but
it's useless with the "wait" instruction.

So long story short: I need either the on-chip 32kHz timer OR the r4k
timer if the 32kHz one is unusable, but not both, and r4k timer is useless
when au1k_idle is in use.

The current in-kernel Alchemy boards all work with the 32kHz timer, so I'm
not against removing R4K_LIB symbols.

Signed-off-by: Steven J. Hill <[email protected]>
Signed-off-by: Ralf Baechle <[email protected]>
/linux-6.15/arch/mips/kernel/
H A DMakefiled7ea335c Thu Nov 15 05:34:17 GMT 2012 Steven J. Hill <[email protected]> MIPS: Remove usage of CSRC_R4K_LIB config option.

Manuel Lauss <[email protected]> writes:

I introduced it as a fallback because early revisions of Alchemy hardware
we shipped had a non-functional 32kHz timer and had to rely on the r4k
timer instead. Previously the r4k timer was initialized regardless, but
it's useless with the "wait" instruction.

So long story short: I need either the on-chip 32kHz timer OR the r4k
timer if the 32kHz one is unusable, but not both, and r4k timer is useless
when au1k_idle is in use.

The current in-kernel Alchemy boards all work with the 32kHz timer, so I'm
not against removing R4K_LIB symbols.

Signed-off-by: Steven J. Hill <[email protected]>
Signed-off-by: Ralf Baechle <[email protected]>
/linux-6.15/arch/mips/
H A DKconfigd7ea335c Thu Nov 15 05:34:17 GMT 2012 Steven J. Hill <[email protected]> MIPS: Remove usage of CSRC_R4K_LIB config option.

Manuel Lauss <[email protected]> writes:

I introduced it as a fallback because early revisions of Alchemy hardware
we shipped had a non-functional 32kHz timer and had to rely on the r4k
timer instead. Previously the r4k timer was initialized regardless, but
it's useless with the "wait" instruction.

So long story short: I need either the on-chip 32kHz timer OR the r4k
timer if the 32kHz one is unusable, but not both, and r4k timer is useless
when au1k_idle is in use.

The current in-kernel Alchemy boards all work with the 32kHz timer, so I'm
not against removing R4K_LIB symbols.

Signed-off-by: Steven J. Hill <[email protected]>
Signed-off-by: Ralf Baechle <[email protected]>