| Message ID | 20201026112222.56894-14-tony@atomide.com |
|---|---|
| State | Superseded |
| Headers | show |
| Series | Drop remaining pdata for am335x and use genpd | expand |
Hi Tony, On Wed, Jan 13, 2021 at 5:23 PM Tony Lindgren <tony@atomide.com> wrote: > * Geert Uytterhoeven <geert@linux-m68k.org> [210113 16:09]: > > On Mon, Oct 26, 2020 at 1:57 PM Tony Lindgren <tony@atomide.com> wrote: > > > We can now enable simple-pm-bus to use genpd. > > > > > > Signed-off-by: Tony Lindgren <tony@atomide.com> > > > > Thanks for your patch, which is now commit 5a230524f87926f2 ("ARM: > > dts: Use simple-pm-bus for genpd for am3 l4_wkup") in v5.11-rc1. > > > > I have bisected a boot failure on BeagleBone Black to this commit. > > Reverting this commit on top of a v5.11-rc3-based tree does not help. > > > > With "earlycon keep_bootcon", I do get some output, the last line is > > > > l4-wkup-clkctrl:00d4:0: failed to disable > > > > I have attached the kernel output of the previous (good) commit, > > and the failing one. > > > > Do you have a clue? > > Strange, bbb has been working for me with NFSroot, that's with > omap2plus_defconfig. Care to send your .config so I can give it a > try tomorrow? Sent by private email. Just tried omap2plus_defconfig, and it also hangs during boot. Gr{oetje,eeting}s, Geert
* Tony Lindgren <tony@atomide.com> [210113 16:48]: > * Geert Uytterhoeven <geert@linux-m68k.org> [210113 16:42]: > > On Wed, Jan 13, 2021 at 5:23 PM Tony Lindgren <tony@atomide.com> wrote: > > > Strange, bbb has been working for me with NFSroot, that's with > > > omap2plus_defconfig. Care to send your .config so I can give it a > > > try tomorrow? > > > > Sent by private email. > > OK thanks. > > > Just tried omap2plus_defconfig, and it also hangs during boot. Hmm and I just tried and bbb still boots NFSroot for me at commit 5a230524f879 ("ARM: dts: Use simple-pm-bus for genpd for am3 l4_wkup"). > Hmm OK. Will give it a try tomorrow. Later, Tony
* Tony Lindgren <tony@atomide.com> [210113 17:30]: > * Tony Lindgren <tony@atomide.com> [210113 16:48]: > > * Geert Uytterhoeven <geert@linux-m68k.org> [210113 16:42]: > > > On Wed, Jan 13, 2021 at 5:23 PM Tony Lindgren <tony@atomide.com> wrote: > > > > Strange, bbb has been working for me with NFSroot, that's with > > > > omap2plus_defconfig. Care to send your .config so I can give it a > > > > try tomorrow? > > > > > > Sent by private email. > > > > OK thanks. > > > > > Just tried omap2plus_defconfig, and it also hangs during boot. > > Hmm and I just tried and bbb still boots NFSroot for me at commit > 5a230524f879 ("ARM: dts: Use simple-pm-bus for genpd for am3 l4_wkup"). > > > Hmm OK. Will give it a try tomorrow. Not sure why your bbb would produce all the deferred probe while mine won't. However, I've noticed that we need the following patch on omap4/5 and dra7 when playing with updating them to use genpd. Maybe this also fixes your boot issue if you can give it a try? Regards, Tony 8< ------------------- diff --git a/drivers/bus/ti-sysc.c b/drivers/bus/ti-sysc.c --- a/drivers/bus/ti-sysc.c +++ b/drivers/bus/ti-sysc.c @@ -642,6 +642,51 @@ static int sysc_parse_and_check_child_range(struct sysc *ddata) return 0; } +/* Interconnect instances to probe before l4_per instances */ +static struct resource early_bus_ranges[] = { + /* am3/4 l4_wkup */ + { .start = 0x44c00000, .end = 0x44c00000 + 0x300000, }, + /* omap4/5 and dra7 l4_cfg */ + { .start = 0x4a000000, .end = 0x4a000000 + 0x300000, }, + /* omap4 l4_wkup */ + { .start = 0x4a300000, .end = 0x4a300000 + 0x30000, }, + /* omap5 and dra7 l4_wkup without dra7 dcan segment */ + { .start = 0x4ae00000, .end = 0x4ae00000 + 0x30000, }, +}; + +static atomic_t sysc_defer = ATOMIC_INIT(10); + +/** + * sysc_defer_non_critical - defer non_critical interconnect probing + * @ddata: device driver data + * + * We want to probe l4_cfg and l4_wkup interconnect instances before any + * l4_per instances as l4_per instances depend on resources on l4_cfg and + * l4_wkup interconnects. + */ +static int sysc_defer_non_critical(struct sysc *ddata) +{ + struct resource *res; + int i; + + if (!atomic_read(&sysc_defer)) + return 0; + + for (i = 0; i < ARRAY_SIZE(early_bus_ranges); i++) { + res = &early_bus_ranges[i]; + if (ddata->module_pa >= res->start && + ddata->module_pa <= res->end) { + atomic_set(&sysc_defer, 0); + + return 0; + } + } + + atomic_dec_if_positive(&sysc_defer); + + return -EPROBE_DEFER; +} + static struct device_node *stdout_path; static void sysc_init_stdout_path(struct sysc *ddata) @@ -870,6 +915,10 @@ static int sysc_map_and_check_registers(struct sysc *ddata) if (error) return error; + error = sysc_defer_non_critical(ddata); + if (error) + return error; + sysc_check_children(ddata); error = sysc_parse_registers(ddata);
Morning Tony & All, On Thu, 2021-01-14 at 09:20 +0200, Tony Lindgren wrote: > * Tony Lindgren <tony@atomide.com> [210113 17:30]: > > * Tony Lindgren <tony@atomide.com> [210113 16:48]: > > > * Geert Uytterhoeven <geert@linux-m68k.org> [210113 16:42]: > > > > On Wed, Jan 13, 2021 at 5:23 PM Tony Lindgren <tony@atomide.com > > > > > wrote: > > > > > Strange, bbb has been working for me with NFSroot, that's > > > > > with > > > > > omap2plus_defconfig. Care to send your .config so I can give > > > > > it a > > > > > try tomorrow? > > > > > > > > Sent by private email. > > > > > > OK thanks. > > > > > > > Just tried omap2plus_defconfig, and it also hangs during boot. > > > > Hmm and I just tried and bbb still boots NFSroot for me at commit > > 5a230524f879 ("ARM: dts: Use simple-pm-bus for genpd for am3 > > l4_wkup"). > > > > > Hmm OK. Will give it a try tomorrow. > > Not sure why your bbb would produce all the deferred probe while mine > won't. > > However, I've noticed that we need the following patch on omap4/5 and > dra7 when playing with updating them to use genpd. > > Maybe this also fixes your boot issue if you can give it a try? > I do also experience the boot problem described by Geert. Head being: b0625afe305253d0831af9289b37c906c18a781b ARM: OMAP2+: Drop legacy platform data for am3 mpuss => BBB boots Ok. Head being: 5a230524f87926f24d637fe62fd689f7f86f5036 ARM: dts: Use simple-pm-bus for genpd for am3 l4_wkup => BBB boot fails. (I don't have any debug parameters => last thing I see in UART is from uBoot: Bytes transferred = 89849 (15ef9 hex) Kernel image @ 0x81000000 [ 0x000000 - 0x928200 ] ## Flattened Device Tree blob at 82000000 Booting using the fdt blob at 0x82000000 Using Device Tree in place at 82000000, end 82018ef8 Starting kernel ... Head being 5a230524f87926f24d637fe62fd689f7f86f5036 + this patch ARM: dts: Use simple-pm-bus for genpd for am3 l4_wkup (+ patch) Boot still fails for me. Please note - my environment takes the intermediate build result file: arch/arm/boot/dts/.am335x-boneblack.dtb.dts.tmp and compiles this with '-@' to allow further overlays. Best Regards Matti Vaittinen
* Vaittinen, Matti <Matti.Vaittinen@fi.rohmeurope.com> [210114 07:58]: > I do also experience the boot problem described by Geert. > > Head being: > b0625afe305253d0831af9289b37c906c18a781b > ARM: OMAP2+: Drop legacy platform data for am3 mpuss > > => BBB boots Ok. > > Head being: > 5a230524f87926f24d637fe62fd689f7f86f5036 > ARM: dts: Use simple-pm-bus for genpd for am3 l4_wkup > > => BBB boot fails. (I don't have any debug parameters => last thing I > see in UART is from uBoot: Looking at the .config Geert sent me, at least CONFIG_SIMPLE_PM_BUS=y is now needed, and probably should be selected or default y for omaps. Adding that removes the -512 errors at least that I started seeing with Geert's config too. Can you guys check if adding CONFIG_SIMPLE_PM_BUS=y fixes the problem? Regards, Tony
Thanks Tony, On Thu, 2021-01-14 at 10:08 +0200, Tony Lindgren wrote: > * Vaittinen, Matti <Matti.Vaittinen@fi.rohmeurope.com> [210114 > 07:58]: > > I do also experience the boot problem described by Geert. > > > > Head being: > > b0625afe305253d0831af9289b37c906c18a781b > > ARM: OMAP2+: Drop legacy platform data for am3 mpuss > > > > => BBB boots Ok. > > > > Head being: > > 5a230524f87926f24d637fe62fd689f7f86f5036 > > ARM: dts: Use simple-pm-bus for genpd for am3 l4_wkup > > > > => BBB boot fails. (I don't have any debug parameters => last thing > > I > > see in UART is from uBoot: > > Looking at the .config Geert sent me, at least CONFIG_SIMPLE_PM_BUS=y > is now needed, and probably should be selected or default y for > omaps. > > Adding that removes the -512 errors at least that I started seeing > with > Geert's config too. > > Can you guys check if adding CONFIG_SIMPLE_PM_BUS=y fixes the > problem? After enabling CONFIG_SIMPLE_PM_BUS 5a230524f87926f24d637fe62fd689f7f86f5036 ARM: dts: Use simple-pm-bus for genpd for am3 l4_wkup boots Ok :) Sounds like it should be enabled on BBB by default. Best Regards Matti Vaittinen
Hi Tony, On Thu, Jan 14, 2021 at 9:08 AM Tony Lindgren <tony@atomide.com> wrote: > * Vaittinen, Matti <Matti.Vaittinen@fi.rohmeurope.com> [210114 07:58]: > > I do also experience the boot problem described by Geert. > > > > Head being: > > b0625afe305253d0831af9289b37c906c18a781b > > ARM: OMAP2+: Drop legacy platform data for am3 mpuss > > > > => BBB boots Ok. > > > > Head being: > > 5a230524f87926f24d637fe62fd689f7f86f5036 > > ARM: dts: Use simple-pm-bus for genpd for am3 l4_wkup > > > > => BBB boot fails. (I don't have any debug parameters => last thing I > > see in UART is from uBoot: > > Looking at the .config Geert sent me, at least CONFIG_SIMPLE_PM_BUS=y > is now needed, and probably should be selected or default y for omaps. > > Adding that removes the -512 errors at least that I started seeing with > Geert's config too. > > Can you guys check if adding CONFIG_SIMPLE_PM_BUS=y fixes the problem? Thanks, that did the trick! Sorry for not realizing that myself. Gr{oetje,eeting}s, Geert
* Geert Uytterhoeven <geert@linux-m68k.org> [210114 08:52]: > On Thu, Jan 14, 2021 at 9:08 AM Tony Lindgren <tony@atomide.com> wrote: > > Can you guys check if adding CONFIG_SIMPLE_PM_BUS=y fixes the problem? > > Thanks, that did the trick! > Sorry for not realizing that myself. OK thanks for testing, will send out a patch soon with both of you in Cc. Regards, Tony
diff --git a/arch/arm/boot/dts/am33xx-l4.dtsi b/arch/arm/boot/dts/am33xx-l4.dtsi --- a/arch/arm/boot/dts/am33xx-l4.dtsi +++ b/arch/arm/boot/dts/am33xx-l4.dtsi @@ -1,5 +1,8 @@ &l4_wkup { /* 0x44c00000 */ - compatible = "ti,am33xx-l4-wkup", "simple-bus"; + compatible = "ti,am33xx-l4-wkup", "simple-pm-bus"; + power-domains = <&prm_wkup>; + clocks = <&l4_wkup_clkctrl AM3_L4_WKUP_L4_WKUP_CLKCTRL 0>; + clock-names = "fck"; reg = <0x44c00000 0x800>, <0x44c00800 0x800>, <0x44c01000 0x400>, @@ -12,7 +15,7 @@ &l4_wkup { /* 0x44c00000 */ <0x00200000 0x44e00000 0x100000>; /* segment 2 */ segment@0 { /* 0x44c00000 */ - compatible = "simple-bus"; + compatible = "simple-pm-bus"; #address-cells = <1>; #size-cells = <1>; ranges = <0x00000000 0x00000000 0x000800>, /* ap 0 */ @@ -22,7 +25,7 @@ segment@0 { /* 0x44c00000 */ }; segment@100000 { /* 0x44d00000 */ - compatible = "simple-bus"; + compatible = "simple-pm-bus"; #address-cells = <1>; #size-cells = <1>; ranges = <0x00000000 0x00100000 0x004000>, /* ap 4 */ @@ -54,7 +57,7 @@ wkup_m3: cpu@0 { }; segment@200000 { /* 0x44e00000 */ - compatible = "simple-bus"; + compatible = "simple-pm-bus"; #address-cells = <1>; #size-cells = <1>; ranges = <0x00000000 0x00200000 0x002000>, /* ap 8 */
We can now enable simple-pm-bus to use genpd. Signed-off-by: Tony Lindgren <tony@atomide.com> --- arch/arm/boot/dts/am33xx-l4.dtsi | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-)