| Message ID | 1448605387-2043-1-git-send-email-ruscur@russell.cc (mailing list archive) |
|---|---|
| State | Accepted |
| Headers | show |
On Fri, 2015-27-11 at 06:23:07 UTC, Russell Currey wrote: > On BMC machines, console output is controlled by the OPAL firmware and is > only flushed when its pollers are called. When the kernel is in a panic > state, it no longer calls these pollers and thus console output does not > completely flush, causing some output from the panic to be lost. > > Output is only actually lost when the kernel is configured to not power off > or reboot after panic (i.e. CONFIG_PANIC_TIMEOUT is set to 0) since OPAL > flushes the console buffer as part of its power down routines. Before this > patch, however, only partial output would be printed during the timeout wait. > > This patch adds a new kmsg_dumper which gets called at panic time to ensure > panic output is not lost. It accomplishes this by calling OPAL_CONSOLE_FLUSH > in the OPAL API, and if that is not available, the pollers are called enough > times to (hopefully) completely flush the buffer. > > The flushing mechanism will only affect output printed at and before the > kmsg_dump call in kernel/panic.c:panic(). As such, the "end Kernel panic" > message may still be truncated as follows: > > >Call Trace: > >[c000000f1f603b00] [c0000000008e9458] dump_stack+0x90/0xbc (unreliable) > >[c000000f1f603b30] [c0000000008e7e78] panic+0xf8/0x2c4 > >[c000000f1f603bc0] [c000000000be4860] mount_block_root+0x288/0x33c > >[c000000f1f603c80] [c000000000be4d14] prepare_namespace+0x1f4/0x254 > >[c000000f1f603d00] [c000000000be43e8] kernel_init_freeable+0x318/0x350 > >[c000000f1f603dc0] [c00000000000bd74] kernel_init+0x24/0x130 > >[c000000f1f603e30] [c0000000000095b0] ret_from_kernel_thread+0x5c/0xac > >---[ end Kernel panic - not > > This functionality is implemented as a kmsg_dumper as it seems to be the > most sensible way to introduce platform-specific functionality to the > panic function. > > Signed-off-by: Russell Currey <ruscur@russell.cc> > Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> Applied to powerpc next, thanks. https://git.kernel.org/powerpc/c/affddff69c55eb68969448f35f cheers
Michael Ellerman <mpe@ellerman.id.au> writes: > On Fri, 2015-27-11 at 06:23:07 UTC, Russell Currey wrote: >> On BMC machines, console output is controlled by the OPAL firmware and is >> only flushed when its pollers are called. When the kernel is in a panic >> state, it no longer calls these pollers and thus console output does not >> completely flush, causing some output from the panic to be lost. >> >> Output is only actually lost when the kernel is configured to not power off >> or reboot after panic (i.e. CONFIG_PANIC_TIMEOUT is set to 0) since OPAL >> flushes the console buffer as part of its power down routines. Before this >> patch, however, only partial output would be printed during the timeout wait. >> >> This patch adds a new kmsg_dumper which gets called at panic time to ensure >> panic output is not lost. It accomplishes this by calling OPAL_CONSOLE_FLUSH >> in the OPAL API, and if that is not available, the pollers are called enough >> times to (hopefully) completely flush the buffer. >> >> The flushing mechanism will only affect output printed at and before the >> kmsg_dump call in kernel/panic.c:panic(). As such, the "end Kernel panic" >> message may still be truncated as follows: >> >> >Call Trace: >> >[c000000f1f603b00] [c0000000008e9458] dump_stack+0x90/0xbc (unreliable) >> >[c000000f1f603b30] [c0000000008e7e78] panic+0xf8/0x2c4 >> >[c000000f1f603bc0] [c000000000be4860] mount_block_root+0x288/0x33c >> >[c000000f1f603c80] [c000000000be4d14] prepare_namespace+0x1f4/0x254 >> >[c000000f1f603d00] [c000000000be43e8] kernel_init_freeable+0x318/0x350 >> >[c000000f1f603dc0] [c00000000000bd74] kernel_init+0x24/0x130 >> >[c000000f1f603e30] [c0000000000095b0] ret_from_kernel_thread+0x5c/0xac >> >---[ end Kernel panic - not >> >> This functionality is implemented as a kmsg_dumper as it seems to be the >> most sensible way to introduce platform-specific functionality to the >> panic function. >> >> Signed-off-by: Russell Currey <ruscur@russell.cc> >> Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> > > Applied to powerpc next, thanks. > > https://git.kernel.org/powerpc/c/affddff69c55eb68969448f35f The firmware interface changed slightly since this kernel patch[1], it added a parameter to OPAL_CONSOLE_FLUSH which accepted the terminal number to flush, theoretically allowing this to be plumbed into TTY layer or something too. So, we'll either have to update this patch or replace it with an updated one. [1] i'm pushing the accepted skiboot patch now.
On Tue, 2016-01-12 at 14:44 +1100, Stewart Smith wrote: > Michael Ellerman <mpe@ellerman.id.au> writes: > > On Fri, 2015-27-11 at 06:23:07 UTC, Russell Currey wrote: > > > On BMC machines, console output is controlled by the OPAL firmware and is > > > only flushed when its pollers are called. When the kernel is in a panic > > > state, it no longer calls these pollers and thus console output does not > > > completely flush, causing some output from the panic to be lost. > > > > > > Output is only actually lost when the kernel is configured to not power > > > off > > > or reboot after panic (i.e. CONFIG_PANIC_TIMEOUT is set to 0) since OPAL > > > flushes the console buffer as part of its power down routines. Before > > > this > > > patch, however, only partial output would be printed during the timeout > > > wait. > > > > > > This patch adds a new kmsg_dumper which gets called at panic time to > > > ensure > > > panic output is not lost. It accomplishes this by calling > > > OPAL_CONSOLE_FLUSH > > > in the OPAL API, and if that is not available, the pollers are called > > > enough > > > times to (hopefully) completely flush the buffer. > > > > > > The flushing mechanism will only affect output printed at and before the > > > kmsg_dump call in kernel/panic.c:panic(). As such, the "end Kernel > > > panic" > > > message may still be truncated as follows: > > > > > > > Call Trace: > > > > [c000000f1f603b00] [c0000000008e9458] dump_stack+0x90/0xbc (unreliable) > > > > [c000000f1f603b30] [c0000000008e7e78] panic+0xf8/0x2c4 > > > > [c000000f1f603bc0] [c000000000be4860] mount_block_root+0x288/0x33c > > > > [c000000f1f603c80] [c000000000be4d14] prepare_namespace+0x1f4/0x254 > > > > [c000000f1f603d00] [c000000000be43e8] kernel_init_freeable+0x318/0x350 > > > > [c000000f1f603dc0] [c00000000000bd74] kernel_init+0x24/0x130 > > > > [c000000f1f603e30] [c0000000000095b0] ret_from_kernel_thread+0x5c/0xac > > > > ---[ end Kernel panic - not > > > > > > This functionality is implemented as a kmsg_dumper as it seems to be the > > > most sensible way to introduce platform-specific functionality to the > > > panic function. > > > > > > Signed-off-by: Russell Currey <ruscur@russell.cc> > > > Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> > > > > Applied to powerpc next, thanks. > > > > https://git.kernel.org/powerpc/c/affddff69c55eb68969448f35f > > The firmware interface changed slightly since this kernel patch[1], it > added a parameter to OPAL_CONSOLE_FLUSH which accepted the terminal > number to flush, theoretically allowing this to be plumbed into TTY > layer or something too. > > So, we'll either have to update this patch or replace it with an updated > one. > > [1] i'm pushing the accepted skiboot patch now. > I'm working on an updated kernel patch to use the new parameter and additional return values, so I suppose it's up to mpe whether or not this patch gets merged now and another gets sent later to amend it, or if this patch gets reverted in next and I can send a V4 adding the new stuff.
On Tue, 2016-01-12 at 15:17 +1100, Russell Currey wrote: > On Tue, 2016-01-12 at 14:44 +1100, Stewart Smith wrote: > > Michael Ellerman <mpe@ellerman.id.au> writes: > > > On Fri, 2015-27-11 at 06:23:07 UTC, Russell Currey wrote: > > > > On BMC machines, console output is controlled by the OPAL firmware and is > > > > only flushed when its pollers are called. When the kernel is in a panic > > > > state, it no longer calls these pollers and thus console output does not > > > > completely flush, causing some output from the panic to be lost. > > > > > > > > Output is only actually lost when the kernel is configured to not power > > > > off > > > > or reboot after panic (i.e. CONFIG_PANIC_TIMEOUT is set to 0) since OPAL > > > > flushes the console buffer as part of its power down routines. Before > > > > this > > > > patch, however, only partial output would be printed during the timeout > > > > wait. > > > > > > > > This patch adds a new kmsg_dumper which gets called at panic time to > > > > ensure > > > > panic output is not lost. It accomplishes this by calling > > > > OPAL_CONSOLE_FLUSH > > > > in the OPAL API, and if that is not available, the pollers are called > > > > enough > > > > times to (hopefully) completely flush the buffer. > > > > > > > > The flushing mechanism will only affect output printed at and before the > > > > kmsg_dump call in kernel/panic.c:panic(). As such, the "end Kernel > > > > panic" > > > > message may still be truncated as follows: > > > > > > > > > Call Trace: > > > > > [c000000f1f603b00] [c0000000008e9458] dump_stack+0x90/0xbc (unreliable) > > > > > [c000000f1f603b30] [c0000000008e7e78] panic+0xf8/0x2c4 > > > > > [c000000f1f603bc0] [c000000000be4860] mount_block_root+0x288/0x33c > > > > > [c000000f1f603c80] [c000000000be4d14] prepare_namespace+0x1f4/0x254 > > > > > [c000000f1f603d00] [c000000000be43e8] kernel_init_freeable+0x318/0x350 > > > > > [c000000f1f603dc0] [c00000000000bd74] kernel_init+0x24/0x130 > > > > > [c000000f1f603e30] [c0000000000095b0] ret_from_kernel_thread+0x5c/0xac > > > > > ---[ end Kernel panic - not > > > > > > > > This functionality is implemented as a kmsg_dumper as it seems to be the > > > > most sensible way to introduce platform-specific functionality to the > > > > panic function. > > > > > > > > Signed-off-by: Russell Currey <ruscur@russell.cc> > > > > Reviewed-by: Andrew Donnellan <andrew.donnellan@au1.ibm.com> > > > > > > Applied to powerpc next, thanks. > > > > > > https://git.kernel.org/powerpc/c/affddff69c55eb68969448f35f > > > > The firmware interface changed slightly since this kernel patch[1], it > > added a parameter to OPAL_CONSOLE_FLUSH which accepted the terminal > > number to flush, theoretically allowing this to be plumbed into TTY > > layer or something too. > > > > So, we'll either have to update this patch or replace it with an updated > > one. > > > > [1] i'm pushing the accepted skiboot patch now. > > > I'm working on an updated kernel patch to use the new parameter and additional > return values, so I suppose it's up to mpe whether or not this patch gets > merged now and another gets sent later to amend it, or if this patch gets > reverted in next and I can send a V4 adding the new stuff. Doh. I'd rather not revert it, unless we have to. Basically we're passing junk in r3, which skiboot is expecting to be the terminal number. So running the current kernel code on the updated skiboot shouldn't crash and burn, it just won't actually work the way it's supposed to. So my preference would be just an incremental patch ASAP to fix the kernel to do the right thing with the new interface. cheers
Michael Ellerman <mpe@ellerman.id.au> writes: >> > > Applied to powerpc next, thanks. >> > > >> > > https://git.kernel.org/powerpc/c/affddff69c55eb68969448f35f >> > >> > The firmware interface changed slightly since this kernel patch[1], it >> > added a parameter to OPAL_CONSOLE_FLUSH which accepted the terminal >> > number to flush, theoretically allowing this to be plumbed into TTY >> > layer or something too. >> > >> > So, we'll either have to update this patch or replace it with an updated >> > one. >> > >> > [1] i'm pushing the accepted skiboot patch now. >> > >> I'm working on an updated kernel patch to use the new parameter and additional >> return values, so I suppose it's up to mpe whether or not this patch gets >> merged now and another gets sent later to amend it, or if this patch gets >> reverted in next and I can send a V4 adding the new stuff. > > Doh. I'd rather not revert it, unless we have to. > > Basically we're passing junk in r3, which skiboot is expecting to be the > terminal number. and skiboot will just return OPAL_PARAMETER and the kernel code ignores the return value, so all will be *fine*. It may even work by accident sometimes. > So running the current kernel code on the updated skiboot shouldn't crash and > burn, it just won't actually work the way it's supposed to. Right, it'll just do nothing. > So my preference would be just an incremental patch ASAP to fix the kernel to > do the right thing with the new interface. I see that's merged now, which is great! Even if someone is bisecting back, things will be fine too.
On Fri, 2016-01-15 at 15:59 +1100, Stewart Smith wrote: > Michael Ellerman <mpe@ellerman.id.au> writes: > > So my preference would be just an incremental patch ASAP to fix the kernel to > > do the right thing with the new interface. > > I see that's merged now, which is great! Even if someone is bisecting > back, things will be fine too. Yep. And the bisection window is only 17 commits, so not a major problem in the end. I've added a check to my scripts so that when a patch touches opal-api.h it yells at me to check skiboot. Hopefully that will prevent this case in future. cheers
diff --git a/arch/powerpc/include/asm/opal-api.h b/arch/powerpc/include/asm/opal-api.h index 8374afe..f8faaae 100644 --- a/arch/powerpc/include/asm/opal-api.h +++ b/arch/powerpc/include/asm/opal-api.h @@ -157,7 +157,8 @@ #define OPAL_LEDS_GET_INDICATOR 114 #define OPAL_LEDS_SET_INDICATOR 115 #define OPAL_CEC_REBOOT2 116 -#define OPAL_LAST 116 +#define OPAL_CONSOLE_FLUSH 117 +#define OPAL_LAST 117 /* Device tree flags */ diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h index 8001159..a5fd407 100644 --- a/arch/powerpc/include/asm/opal.h +++ b/arch/powerpc/include/asm/opal.h @@ -35,6 +35,7 @@ int64_t opal_console_read(int64_t term_number, __be64 *length, uint8_t *buffer); int64_t opal_console_write_buffer_space(int64_t term_number, __be64 *length); +void opal_console_flush(void); int64_t opal_rtc_read(__be32 *year_month_day, __be64 *hour_minute_second_millisecond); int64_t opal_rtc_write(uint32_t year_month_day, @@ -262,6 +263,8 @@ extern int opal_resync_timebase(void); extern void opal_lpc_init(void); +extern void opal_kmsg_init(void); + extern int opal_event_request(unsigned int opal_event_nr); struct opal_sg_list *opal_vmalloc_to_sg_list(void *vmalloc_addr, diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile index 1c8cdb6..b9de7ef 100644 --- a/arch/powerpc/platforms/powernv/Makefile +++ b/arch/powerpc/platforms/powernv/Makefile @@ -2,6 +2,7 @@ obj-y += setup.o opal-wrappers.o opal.o opal-async.o idle.o obj-y += opal-rtc.o opal-nvram.o opal-lpc.o opal-flash.o obj-y += rng.o opal-elog.o opal-dump.o opal-sysparam.o opal-sensor.o obj-y += opal-msglog.o opal-hmi.o opal-power.o opal-irqchip.o +obj-y += opal-kmsg.o obj-$(CONFIG_SMP) += smp.o subcore.o subcore-asm.o obj-$(CONFIG_PCI) += pci.o pci-p5ioc2.o pci-ioda.o diff --git a/arch/powerpc/platforms/powernv/opal-kmsg.c b/arch/powerpc/platforms/powernv/opal-kmsg.c new file mode 100644 index 0000000..bd3b2ee --- /dev/null +++ b/arch/powerpc/platforms/powernv/opal-kmsg.c @@ -0,0 +1,68 @@ +/* + * kmsg dumper that ensures the OPAL console fully flushes panic messages + * + * Author: Russell Currey <ruscur@russell.cc> + * + * Copyright 2015 IBM Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + */ + +#include <linux/kmsg_dump.h> + +#include <asm/opal.h> +#include <asm/opal-api.h> + +/* + * Console output is controlled by OPAL firmware. The kernel regularly calls + * OPAL_POLL_EVENTS, which flushes some console output. In a panic state, + * however, the kernel no longer calls OPAL_POLL_EVENTS and the panic message + * may not be completely printed. This function does not actually dump the + * message, it just ensures that OPAL completely flushes the console buffer. + */ +static void force_opal_console_flush(struct kmsg_dumper *dumper, + enum kmsg_dump_reason reason) +{ + int i; + + /* + * Outside of a panic context the pollers will continue to run, + * so we don't need to do any special flushing. + */ + if (reason != KMSG_DUMP_PANIC) + return; + + if (opal_check_token(OPAL_CONSOLE_FLUSH)) { + opal_console_flush(); + } else { + /* + * If OPAL_CONSOLE_FLUSH is not implemented in the firmware, + * the console can still be flushed by calling the polling + * function enough times to flush the buffer. We don't know + * how much output still needs to be flushed, but we can be + * generous since the kernel is in panic and doesn't need + * to do much else. + */ + printk(KERN_NOTICE "opal: OPAL_CONSOLE_FLUSH missing.\n"); + for (i = 0; i < 1024; i++) { + opal_poll_events(NULL); + } + } +} + +static struct kmsg_dumper opal_kmsg_dumper = { + .dump = force_opal_console_flush +}; + +void __init opal_kmsg_init(void) +{ + int rc; + + /* Add our dumper to the list */ + rc = kmsg_dump_register(&opal_kmsg_dumper); + if (rc != 0) + pr_err("opal: kmsg_dump_register failed; returned %d\n", rc); +} diff --git a/arch/powerpc/platforms/powernv/opal-wrappers.S b/arch/powerpc/platforms/powernv/opal-wrappers.S index b7a464f..e45b88a 100644 --- a/arch/powerpc/platforms/powernv/opal-wrappers.S +++ b/arch/powerpc/platforms/powernv/opal-wrappers.S @@ -301,3 +301,4 @@ OPAL_CALL(opal_flash_erase, OPAL_FLASH_ERASE); OPAL_CALL(opal_prd_msg, OPAL_PRD_MSG); OPAL_CALL(opal_leds_get_ind, OPAL_LEDS_GET_INDICATOR); OPAL_CALL(opal_leds_set_ind, OPAL_LEDS_SET_INDICATOR); +OPAL_CALL(opal_console_flush, OPAL_CONSOLE_FLUSH); diff --git a/arch/powerpc/platforms/powernv/opal.c b/arch/powerpc/platforms/powernv/opal.c index 4296d55..a94a85c 100644 --- a/arch/powerpc/platforms/powernv/opal.c +++ b/arch/powerpc/platforms/powernv/opal.c @@ -758,6 +758,9 @@ static int __init opal_init(void) opal_pdev_init(opal_node, "ibm,opal-flash"); opal_pdev_init(opal_node, "ibm,opal-prd"); + /* Initialise OPAL kmsg dumper for flushing console on panic */ + opal_kmsg_init(); + return 0; } machine_subsys_initcall(powernv, opal_init);