Hi, On Thu, May 7, 2020 at 1:09 PM Douglas Anderson <dianders@xxxxxxxxxxxx> wrote: > > In order to make early kgdb work properly we need early_brk64() to be > able to call into it. This is as easy as adding a call into > call_break_hook() just like we do later in the normal brk_handler(). > > Once we do this we can let kgdb know that it can break into the > debugger a little earlier (specifically when parsing early_param's). > > NOTE: without this patch it turns out that arm64 can't do breakpoints > even at dbg_late_init(), so if we decide something about this patch is > wrong we might need to move dbg_late_init() a little later. > > Signed-off-by: Douglas Anderson <dianders@xxxxxxxxxxxx> > Cc: Catalin Marinas <catalin.marinas@xxxxxxx> > Cc: Will Deacon <will@xxxxxxxxxx> > Reviewed-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> > Reviewed-by: Daniel Thompson <daniel.thompson@xxxxxxxxxx> > --- > > Changes in v4: > - Add "if KGDB" to "select ARCH_HAS_EARLY_DEBUG" in Kconfig. > > Changes in v3: > - Change boolean weak function to KConfig. > > Changes in v2: None > > arch/arm64/Kconfig | 1 + > arch/arm64/include/asm/debug-monitors.h | 2 ++ > arch/arm64/kernel/debug-monitors.c | 2 +- > arch/arm64/kernel/traps.c | 3 +++ > 4 files changed, 7 insertions(+), 1 deletion(-) As discussed in the replies to the v3 version of this patch [1], I have posted a replacement patch for this one [2]. After the cut in the replacement patch I talk about different ways it could land. Hopefully that's not too confusing. I can also re-spam everyone with a v5 of the whole series if that makes it clearer. [1] https://lore.kernel.org/r/20200428141218.v3.5.I22067ad43e77ddfd4b64c2d49030628480f9e8d9@changeid [2] https://lore.kernel.org/r/20200513160501.1.I0b5edf030cc6ebef6ab4829f8867cdaea42485d8@changeid -Doug