Cleaning printk stuff in NMI context
Subsystems (10)
Petr Mladek
maintainerv4 cover Cleaning printk stuff in NMI context
Hi Andrew,
this is just a quick respin of the previous version. It changes the few details as you suggested. Also it fixes the build problem on ARM as reported by Geert and Arnd.
I rather send the whole patch set because there is the renamed header. Also the extra blank space affects two patches. I hope that it will safe you some work. Please, let me know if you would prefer incremental patches.
Changes against v3:
+ used size_t for "len" and "size"
+ replaced WARN() with pr_err()
+ renamed kernel/printk/printk.h -> internal.h
+ fixed build on ARM (undefined NMI_LOG_BUF_SHIFT)
Changes against v2:
+ fixed compilation problems reported by 0-day build robot
+ MN10300 and Xtensa architectures will get handled separately
+ dropped the patch that printed NMI messages directly when Oops in progress; it made the solution less reliable
+ made the size of the buffer configurable; use real numbers instead of PAGE_SIZE
Changes against v1:
+ rebased on top of 4.4-rc2; there the old implementation was moved to lib/nmi_backtrace.c and used also on arm; I hope that I got the arm side correctly; I was not able to test on arm :-(
+ defined HAVE_NMI on arm for !CPU_V7M instead of !CPU_V7; handle_fiq_as_nmi() is called from entry-armv.S that is compiled when !CPU_V7M
+ defined HAVE_NMI also on mips; it calls nmi_enter() and seems to have real NMIs (or am I wrong?)
+ serialized backtraces when printing directly
(oops_in_progress)
Petr Mladek (4):
printk/nmi: Generic solution for safe printk in NMI printk/nmi: Use IRQ work only when ready
printk/nmi: Warn when some message has been lost in NMI context printk/nmi: Increase the size of NMI buffer and make it configurable
arch/Kconfig | 7 ++ arch/arm/Kconfig | 2 + arch/arm/kernel/smp.c | 2 + arch/avr32/Kconfig | 1 + arch/blackfin/Kconfig | 1 + arch/cris/Kconfig | 1 + arch/mips/Kconfig | 1 + arch/powerpc/Kconfig | 1 + arch/s390/Kconfig | 1 + arch/sh/Kconfig | 1 + arch/sparc/Kconfig | 1 + arch/tile/Kconfig | 1 + arch/x86/Kconfig | 1 + arch/x86/kernel/apic/hw_nmi.c | 1 - include/linux/hardirq.h | 2 + include/linux/percpu.h | 3 - include/linux/printk.h | 12 ++- init/Kconfig | 27 +++++ init/main.c | 1 + kernel/printk/Makefile | 1 + kernel/printk/internal.h | 55 +++++++++++ kernel/printk/nmi.c | 222 ++++++++++++++++++++++++++++++++++++++++++ kernel/printk/printk.c | 29 +++--- lib/nmi_backtrace.c | 89 +---------------- 24 files changed, 356 insertions(+), 107 deletions(-) create mode 100644 kernel/printk/internal.h create mode 100644 kernel/printk/nmi.c
Signature · 1 line
1.8.5.6
Petr Mladek
maintainerPATCH v4 1/4 printk/nmi: Generic solution for safe printk in NMI
printk() takes some locks and could not be used a safe way in NMI context.
The chance of a deadlock is real especially when printing stacks from all CPUs. This particular problem has been addressed on x86 by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs").
This patch reuses most of the code and makes it generic. It is useful for all messages and architectures that support NMI.
The alternative printk_func is set when entering and is reseted when leaving NMI context. It queues IRQ work to copy the messages into the main ring buffer in a safe context.
__printk_nmi_flush() copies all available messages and reset the buffer. Then we could use a simple cmpxchg operations to get synchronized with writers. There is also used a spinlock to get synchronized with other flushers.
We do not longer use seq_buf because it depends on external lock. It would be hard to make all supported operations safe for a lockless use. It would be confusing and error prone to make only some operations safe.
The code is put into separate printk/nmi.c as suggested by Steven Rostedt. It needs a per-CPU buffer and is compiled only on architectures that call nmi_enter(). This is achieved by the new HAVE_NMI Kconfig flag.
One exception is arm where the deferred printing is used for printing backtraces even without NMI. For this purpose, we define NEED_PRINTK_NMI Kconfig flag. The alternative printk_func is explicitly set when IPI_CPU_BACKTRACE is handled.
The other exceptions are MN10300 and Xtensa architectures. We need to clean up NMI handling there first. Let's do it separately.
The patch is heavily based on the draft from Peter Zijlstra, see https://lkml.org/lkml/2015/6/10/327
--- arch/Kconfig | 7 ++ arch/arm/Kconfig | 2 + arch/arm/kernel/smp.c | 2 + arch/avr32/Kconfig | 1 + arch/blackfin/Kconfig | 1 + arch/cris/Kconfig | 1 + arch/mips/Kconfig | 1 + arch/powerpc/Kconfig | 1 + arch/s390/Kconfig | 1 + arch/sh/Kconfig | 1 + arch/sparc/Kconfig | 1 + arch/tile/Kconfig | 1 + arch/x86/Kconfig | 1 + arch/x86/kernel/apic/hw_nmi.c | 1 - include/linux/hardirq.h | 2 + include/linux/percpu.h | 3 - include/linux/printk.h | 12 ++- init/Kconfig | 5 + init/main.c | 1 + kernel/printk/Makefile | 1 + kernel/printk/internal.h | 44 +++++++++ kernel/printk/nmi.c | 207 ++++++++++++++++++++++++++++++++++++++++++ kernel/printk/printk.c | 19 +--- lib/nmi_backtrace.c | 89 +----------------- 24 files changed, 298 insertions(+), 107 deletions(-) create mode 100644 kernel/printk/internal.h create mode 100644 kernel/printk/nmi.c
arch/Kconfig +7 -0
@@ -187,7 +187,14 @@ config HAVE_OPTPROBES config HAVE_KPROBES_ON_FTRACE bool +config HAVE_NMI + bool + +config NEED_PRINTK_NMI + bool + config HAVE_NMI_WATCHDOG + depends on HAVE_NMI bool # # An arch should select this if it provides all these things:
arch/arm/Kconfig +2 -0
@@ -63,6 +63,8 @@ config ARM select HAVE_KRETPROBES if (HAVE_KPROBES) select HAVE_MEMBLOCK select HAVE_MOD_ARCH_SPECIFIC + select HAVE_NMI if (!CPU_V7M) + select NEED_PRINTK_NMI if (CPU_V7M) select HAVE_OPROFILE if (HAVE_PERF_EVENTS) select HAVE_OPTPROBES if !THUMB2_KERNEL select HAVE_PERF_EVENTS
arch/arm/kernel/smp.c +2 -0
@@ -648,7 +648,9 @@ void handle_IPI(int ipinr, struct pt_regs *regs)case IPI_CPU_BACKTRACE :irq_enter (); +printk_nmi_enter ();nmi_cpu_backtrace (regs); +printk_nmi_exit ();irq_exit ();break ;
arch/avr32/Kconfig +1 -0
@@ -17,6 +17,7 @@ config AVR32 select GENERIC_CLOCKEVENTS select HAVE_MOD_ARCH_SPECIFIC select MODULES_USE_ELF_RELA + select HAVE_NMI help AVR32 is a high-performance 32-bit RISC microprocessor core, designed for cost-sensitive embedded applications, with particular
arch/blackfin/Kconfig +1 -0
@@ -40,6 +40,7 @@ config BLACKFIN select HAVE_MOD_ARCH_SPECIFIC select MODULES_USE_ELF_RELA select HAVE_DEBUG_STACKOVERFLOW + select HAVE_NMI config GENERIC_CSUM def_bool y
arch/cris/Kconfig +1 -0
@@ -69,6 +69,7 @@ config CRIS select GENERIC_CLOCKEVENTS if ETRAX_ARCH_V32 select GENERIC_SCHED_CLOCK if ETRAX_ARCH_V32 select HAVE_DEBUG_BUGVERBOSE if ETRAX_ARCH_V32 + select HAVE_NMI config HZ int
arch/mips/Kconfig +1 -0
@@ -63,6 +63,7 @@ config MIPS select HAVE_IRQ_TIME_ACCOUNTING select GENERIC_TIME_VSYSCALL select ARCH_CLOCKSOURCE_DATA + select HAVE_NMI menu "Machine selection"
arch/powerpc/Kconfig +1 -0
@@ -156,6 +156,7 @@ config PPC select NO_BOOTMEM select HAVE_GENERIC_RCU_GUP select HAVE_PERF_EVENTS_NMI if PPC64 + select HAVE_NMI if PERF_EVENTS select EDAC_SUPPORT select EDAC_ATOMIC_SCRUB select ARCH_HAS_DMA_SET_COHERENT_MASK
arch/s390/Kconfig +1 -0
@@ -159,6 +159,7 @@ config S390 select TTY select VIRT_CPU_ACCOUNTING select VIRT_TO_BUS + select HAVE_NMI config SCHED_OMIT_FRAME_POINTER
arch/sh/Kconfig +1 -0
@@ -44,6 +44,7 @@ config SUPERH select OLD_SIGSUSPEND select OLD_SIGACTION select HAVE_ARCH_AUDITSYSCALL + select HAVE_NMI help The SuperH is a RISC processor targeted for use in embedded systems and consumer electronics; it was also used in the Sega Dreamcast
arch/sparc/Kconfig +1 -0
@@ -80,6 +80,7 @@ config SPARC64 select NO_BOOTMEM select HAVE_ARCH_AUDITSYSCALL select ARCH_SUPPORTS_ATOMIC_RMW + select HAVE_NMI config ARCH_DEFCONFIG string
arch/tile/Kconfig +1 -0
@@ -29,6 +29,7 @@ config TILE select HAVE_DEBUG_STACKOVERFLOW select ARCH_WANT_FRAME_POINTERS select HAVE_CONTEXT_TRACKING + select HAVE_NMI if USE_PMC select EDAC_SUPPORT select GENERIC_STRNCPY_FROM_USER select GENERIC_STRNLEN_USER
arch/x86/Kconfig +1 -0
@@ -126,6 +126,7 @@ config X86 select HAVE_MEMBLOCK select HAVE_MEMBLOCK_NODE_MAP select HAVE_MIXED_BREAKPOINTS_REGS + select HAVE_NMI select HAVE_OPROFILE select HAVE_OPTPROBES select HAVE_PCSPKR_PLATFORM
arch/x86/kernel/apic/hw_nmi.c +0 -1
@@ -18,7 +18,6 @@#include <linux/nmi.h> #include <linux/module.h> #include <linux/delay.h> -#include <linux/seq_buf.h> #ifdef CONFIG_HARDLOCKUP_DETECTOR u64 hw_nmi_get_sample_period (int watchdog_thresh)
include/linux/hardirq.h +2 -0
@@ -67,10 +67,12 @@ extern void irq_exit(void);preempt_count_add (NMI_OFFSET +HARDIRQ_OFFSET ); \rcu_nmi_enter (); \trace_hardirq_enter (); \ +printk_nmi_enter (); \ } while (0)#define nmi_exit () \ do { \ + printk_nmi_exit(); \ trace_hardirq_exit(); \ rcu_nmi_exit(); \ BUG_ON(!in_nmi()); \
include/linux/percpu.h +0 -3
@@ -135,7 +135,4 @@ extern phys_addr_t per_cpu_ptr_to_phys(void *addr); (typeof(type) __percpu *)__alloc_percpu(sizeof(type), \ __alignof__(type)) -/* To avoid include hell, as printk can not declare this, we declare it here */ -DECLARE_PER_CPU (printk_func_t, printk_func); -#endif /* __LINUX_PERCPU_H */
include/linux/printk.h +11 -1
@@ -122,7 +122,17 @@ static inline __printf(1, 2) __coldvoid early_printk (const char *s, ...) { }#endif -typedef __printf(1, 0) int (*printk_func_t)(const char *fmt, va_list args); +#ifdef CONFIG_PRINTK_NMI +extern void printk_nmi_init (void ); +extern void printk_nmi_enter (void ); +extern void printk_nmi_exit (void ); +extern void printk_nmi_flush (void ); +#else +static inline void printk_nmi_init (void ) { } +static inline void printk_nmi_enter (void ) { } +static inline void printk_nmi_exit (void ) { } +static inline void printk_nmi_flush (void ) { } +#endif /* PRINTK_NMI */ #ifdef CONFIG_PRINTK asmlinkage __printf(5, 0)
init/Kconfig +5 -0
@@ -1456,6 +1456,11 @@ config PRINTK very difficult to diagnose system problems, saying N here is strongly discouraged. +config PRINTK_NMI + def_bool y + depends on PRINTK + depends on HAVE_NMI || NEED_PRINTK_NMI + config BUG bool "BUG() support" if EXPERT default y
init/main.c +1 -0
@@ -591,6 +591,7 @@ asmlinkage __visible void __init start_kernel(void)timekeeping_init ();time_init ();sched_clock_postinit (); +printk_nmi_init ();perf_event_init ();profile_init ();call_function_init ();
kernel/printk/Makefile +1 -0
@@ -1,2 +1,3 @@ obj-y = printk.o +obj-$(CONFIG_PRINTK_NMI) += nmi.o obj-$(CONFIG_A11Y_BRAILLE_CONSOLE) += braille.o
kernel/printk/internal.h new file +44 -0
44 lines not on this page · raw message
kernel/printk/nmi.c new file +207 -0
207 lines not on this page · raw message
kernel/printk/printk.c +1 -18
@@ -54,6 +54,7 @@#include "console_cmdline.h" #include "braille.h" +#include "internal.h" int console_printk[4 ] = {CONSOLE_LOGLEVEL_DEFAULT ,/* console_loglevel */
@@ -1867,14 +1868,6 @@ int vprintk_default(const char *fmt, va_list args) }EXPORT_SYMBOL_GPL (vprintk_default); -/* -* This allows printk to be diverted to another function per cpu. -* This is useful for calling printk functions from within NMI -* without worrying about race conditions that can lock up the -* box. -*/ -DEFINE_PER_CPU (printk_func_t, printk_func) = vprintk_default; -/** * printk - print a kernel message * @fmt: format string
@@ -1898,21 +1891,11 @@ DEFINE_PER_CPU(printk_func_t, printk_func) = vprintk_default;*/ asmlinkage __visible int printk(const char *fmt, ...) { -printk_func_t vprintk_func;va_list args;int r;va_start (args, fmt); - -/* -* If a caller overrides the per_cpu printk_func, then it needs -* to disable preemption when calling printk(). Otherwise -* the printk_func should be set to the default. No need to -* disable preemption here. -*/ - vprintk_func =this_cpu_read (printk_func); r =vprintk_func (fmt, args); -va_end (args);return r;
lib/nmi_backtrace.c +5 -84
136 lines not on this page · raw message
Signature · 1 line
1.8.5.6
Petr Mladek
maintainerPATCH v4 2/4 printk/nmi: Use IRQ work only when ready
In reply to Petr MladekNMIs could happen at any time. This patch makes sure that the safe printk() in NMI will schedule IRQ work only when the related structs are initialized.
All pending messages are flushed when the IRQ work is being initialized.
--- kernel/printk/nmi.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)
kernel/printk/nmi.c +12 -1
@@ -38,6 +38,7 @@* were handled or when IRQs are blocked. */ DEFINE_PER_CPU (printk_func_t, printk_func) = vprintk_default; +static int printk_nmi_irq_ready;#define NMI_LOG_BUF_LEN (4096 - sizeof(atomic_t) - sizeof(struct irq_work))
@@ -84,8 +85,11 @@ again: goto again;/* Get flushed in a more safe context. */ -if (add) +if (add && printk_nmi_irq_ready) { +/* Make sure that IRQ work is really initialized. */ +smp_rmb ();irq_work_queue (&s->work); + }return add; }
@@ -194,6 +198,13 @@ void __init printk_nmi_init(void)init_irq_work (&s->work, __printk_nmi_flush); } + +/* Make sure that IRQ works are initialized before enabling. */ +smp_wmb (); + printk_nmi_irq_ready =1 ; + +/* Flush pending messages that did not have scheduled IRQ works. */ +printk_nmi_flush (); }void printk_nmi_enter (void )
Signature · 1 line
1.8.5.6
Petr Mladek
maintainerPATCH v4 3/4 printk/nmi: Warn when some message has been lost in NMI context
In reply to Petr MladekWe could not resize the temporary buffer in NMI context. Let's warn if a message is lost.
This is rather theoretical. printk() should not be used in NMI. The only sensible use is when we want to print backtrace from all CPUs. The current buffer should be enough for this purpose.
--- kernel/printk/internal.h | 11 +++++++++++ kernel/printk/nmi.c | 5 ++++- kernel/printk/printk.c | 10 ++++++++++ 3 files changed, 25 insertions(+), 1 deletion(-)
kernel/printk/internal.h +11 -0
@@ -34,6 +34,12 @@ static inline __printf(1, 0) int vprintk_func(const char *fmt, va_list args)return this_cpu_read (printk_func)(fmt, args); } +extern atomic_t nmi_message_lost; +static inline int get_nmi_message_lost (void ) +{ +return atomic_xchg (&nmi_message_lost,0 ); +} +#else /* CONFIG_PRINTK_NMI */ static inline __printf(1, 0) int vprintk_func(const char *fmt, va_list args)
@@ -41,4 +47,9 @@ static inline __printf(1, 0) int vprintk_func(const char *fmt, va_list args) return vprintk_default(fmt, args); } +static inline int get_nmi_message_lost (void ) +{ +return 0 ; +} +#endif /* CONFIG_PRINTK_NMI */
kernel/printk/nmi.c +4 -1
@@ -39,6 +39,7 @@*/ DEFINE_PER_CPU (printk_func_t, printk_func) = vprintk_default;static int printk_nmi_irq_ready; +atomic_t nmi_message_lost;#define NMI_LOG_BUF_LEN (4096 - sizeof(atomic_t) - sizeof(struct irq_work))
@@ -64,8 +65,10 @@ static int vprintk_nmi(const char *fmt, va_list args) again: len =atomic_read (&s->len); -if (len >=sizeof (s->buffer)) +if (len >=sizeof (s->buffer)) { +atomic_inc (&nmi_message_lost);return 0 ; + }/* * Make sure that all old data have been read before the buffer was
kernel/printk/printk.c +10 -0
@@ -1669,6 +1669,7 @@ asmlinkage int vprintk_emit(int facility, int level,unsigned long flags;int this_cpu;int printed_len =0 ; +int nmi_message_lost;bool in_sched = false;/* cpu currently holding logbuf_lock in this function */ static unsigned int logbuf_cpu =UINT_MAX ;
@@ -1719,6 +1720,15 @@ asmlinkage int vprintk_emit(int facility, int level, strlen(recursion_msg)); } + nmi_message_lost =get_nmi_message_lost (); +if (unlikely (nmi_message_lost)) { + text_len =scnprintf (textbuf,sizeof (textbuf), +"BAD LUCK: lost %d message(s) from NMI context!" , + nmi_message_lost); + printed_len +=log_store (0 ,2 ,LOG_PREFIX |LOG_NEWLINE ,0 , +NULL ,0 , textbuf, text_len); + } +/* * The printf needs to come first; we need the syslog * prefix which might be passed-in as a parameter.
Signature · 1 line
1.8.5.6
Petr Mladek
maintainerPATCH v4 4/4 printk/nmi: Increase the size of NMI buffer and make it configurable
In reply to Petr MladekTesting has shown that the backtrace sometimes does not fit into the 4kB temporary buffer that is used in NMI context. The warnings are gone when I double the temporary buffer size.
This patch doubles the buffer size and makes it configurable.
Note that this problem existed even in the x86-specific implementation that was added by the commit a9edc8809328 ("x86/nmi: Perform a safe NMI stack trace on all CPUs"). Nobody noticed it because it did not print any warnings.
--- init/Kconfig | 22 ++++++++++++++++++++++ kernel/printk/nmi.c | 3 ++- 2 files changed, 24 insertions(+), 1 deletion(-)
init/Kconfig +22 -0
@@ -866,6 +866,28 @@ config LOG_CPU_MAX_BUF_SHIFT 13 => 8 KB for each CPU 12 => 4 KB for each CPU +config NMI_LOG_BUF_SHIFT + int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)" + range 10 21 + default 13 + depends on PRINTK_NMI + help + Select the size of a per-CPU buffer where NMI messages are temporary + stored. They are copied to the main log buffer in a safe context + to avoid a deadlock. The value defines the size as a power of 2. + + NMI messages are rare and limited. The largest one is when + a backtrace is printed. It usually fits into 4KB. Select + 8KB if you want to be on the safe side. + + Examples: + 17 => 128 KB for each CPU + 16 => 64 KB for each CPU + 15 => 32 KB for each CPU + 14 => 16 KB for each CPU + 13 => 8 KB for each CPU + 12 => 4 KB for each CPU + # # Architectures with an unreliable sched_clock() should select this: #
kernel/printk/nmi.c +2 -1
@@ -41,7 +41,8 @@ DEFINE_PER_CPU(printk_func_t, printk_func) = vprintk_default;static int printk_nmi_irq_ready;atomic_t nmi_message_lost; -#define NMI_LOG_BUF_LEN (4096 - sizeof(atomic_t) - sizeof(struct irq_work)) +#define NMI_LOG_BUF_LEN ((1 << CONFIG_NMI_LOG_BUF_SHIFT) - \ + sizeof(atomic_t) - sizeof(struct irq_work))struct nmi_seq_buf {atomic_t len;/* length of written data */
Signature · 1 line
1.8.5.6
Petr Mladek
maintainerQuoted 14 lines · Petr Mladek
Hi Andrew,
this is just a quick respin of the previous version. It changes the few details as you suggested. Also it fixes the build problem on ARM as reported by Geert and Arnd.
I rather send the whole patch set because there is the renamed header. Also the extra blank space affects two patches. I hope that it will safe you some work. Please, let me know if you would prefer incremental patches.
Changes against v3:
Quoted 7 lines
+ used
size_tfor "len" and "size"+ replaced
WARN()withpr_err()+ renamed
kernel/printk/printk.h->internal.h+ fixed build on ARM (undefined
NMI_LOG_BUF_SHIFT)
Please, find below the diff between the two branches based on v3 and v4 of the patch set. I hope that it might help.
init/Kconfig +1 -1
@@ -870,7 +870,7 @@ config NMI_LOG_BUF_SHIFT int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)" range 10 21 default 13 - depends on PRINTK && HAVE_NMI + depends on PRINTK_NMI help Select the size of a per-CPU buffer where NMI messages are temporary stored. They are copied to the main log buffer in a safe context
kernel/printk/internal.h renamed from kernel/printk/printk.h +1 -1
@@ -1,5 +1,5 @@/* -* printk.h - printk internal definitions +* internal.h - printk internal definitions * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License
kernel/printk/nmi.c +9 -5
@@ -22,7 +22,7 @@#include <linux/irq_work.h> #include <linux/printk.h> -#include "printk.h" +#include "internal.h" /* * printk() could not take logbuf_lock in NMI context. Instead,
@@ -60,12 +60,13 @@ static DEFINE_PER_CPU(struct nmi_seq_buf, nmi_print_seq);static int vprintk_nmi (const char *fmt,va_list args) {struct nmi_seq_buf *s =this_cpu_ptr (&nmi_print_seq); -int add =0 , len; +int add =0 ; +size_t len; again: len =atomic_read (&s->len); -if (len >=sizeof (s->buffer)) { +if (len >=sizeof (s->buffer)) {atomic_inc (&nmi_message_lost);return 0 ; }
@@ -117,7 +118,8 @@ static void __printk_nmi_flush(struct irq_work *work)static raw_spinlock_t read_lock =__RAW_SPIN_LOCK_INITIALIZER (read_lock); struct nmi_seq_buf *s = container_of(work, struct nmi_seq_buf, work); -int len, size, i, last_i; +size_t len, size; +int i, last_i;/* * The lock has two functions. First, one reader has to flush all
@@ -137,7 +139,9 @@ more:* the buffer an unexpected way. If we printed something then * @len must only increase. */ -WARN_ON (i && i >= len); +if (i && i >= len) +pr_err ("printk_nmi_flush: internal error: i=%d >= len=%lu\n" , + i, len);if (!len) goto out;/* Someone else has already flushed the buffer. */
kernel/printk/printk.c +1 -1
@@ -54,7 +54,7 @@#include "console_cmdline.h" #include "braille.h" -#include "printk.h" +#include "internal.h" int console_printk[4 ] = {CONSOLE_LOGLEVEL_DEFAULT ,/* console_loglevel */
Andrew Morton
maintainerQuoted 12 lines · On Fri, 11 Dec 2015 14:20:48 +0100 Petr Mladek <[email protected]> wrote:
this is just a quick respin of the previous version. It changes the few details as you suggested. Also it fixes the build problem on ARM as reported by Geert and Arnd.
I rather send the whole patch set because there is the renamed header. Also the extra blank space affects two patches. I hope that it will safe you some work. Please, let me know if you would prefer incremental patches.
Changes against v3:
Quoted 7 lines
+ used
size_tfor "len" and "size"+ replaced
WARN()withpr_err()+ renamed
kernel/printk/printk.h->internal.h+ fixed build on ARM (undefined
NMI_LOG_BUF_SHIFT)
So the review of the v3 patchset was ... inconclusive. And everyone has gone quiet about v4.
Probably because you didn't cc the V3 discussion participants when sending out V4. Big mistake, sorry, I can't check everything!
But v4 is basically unaltered from v3 so can we please rev this up again? yay or nay? Thanks.
Petr Mladek
maintainerQuoted 20 lines · Andrew Morton
Quoted 18 lines · On Fri, 11 Dec 2015 14:20:48 +0100 Petr Mladek <[email protected]> wrote:
this is just a quick respin of the previous version. It changes the few details as you suggested. Also it fixes the build problem on ARM as reported by Geert and Arnd.
I rather send the whole patch set because there is the renamed header. Also the extra blank space affects two patches. I hope that it will safe you some work. Please, let me know if you would prefer incremental patches.
Changes against v3:
+ used
size_tfor "len" and "size"+ replaced
WARN()withpr_err()+ renamed
kernel/printk/printk.h->internal.h
Quoted 7 lines
Quoted 1 line
+ fixed build on ARM (undefined
NMI_LOG_BUF_SHIFT)So the review of the v3 patchset was ... inconclusive. And everyone has gone quiet about v4.
Probably because you didn't cc the V3 discussion participants when sending out V4. Big mistake, sorry, I can't check everything!
Ah, v4 was sent too fast after v3 before others reacted. It included rather cosmetic changes based on your (Andrew's) feedback.
Quoted 2 lines
But v4 is basically unaltered from v3 so can we please rev this up again? yay or nay? Thanks.
The patch set prevents deadlocks that happen in a real life. It increases a chance to get a valid crash dump when NMI is involved.
It does not handle well the situation when NMI is involved and the crash dump could not get produced. But this is not handled well even now. As Peter Zijlstra writes this a usually a real mess when he needs to use special hacks (early_printk) anyway.
I still believe that the patch set makes sense and is acceptable as is.
Thanks for taking care of it,
Petr
Petr Mladek
maintainerQuoted 39 lines · Petr Mladek
Quoted 20 lines · Andrew Morton
Quoted 18 lines · On Fri, 11 Dec 2015 14:20:48 +0100 Petr Mladek <[email protected]> wrote:
this is just a quick respin of the previous version. It changes the few details as you suggested. Also it fixes the build problem on ARM as reported by Geert and Arnd.
I rather send the whole patch set because there is the renamed header. Also the extra blank space affects two patches. I hope that it will safe you some work. Please, let me know if you would prefer incremental patches.
Changes against v3:
+ used
size_tfor "len" and "size"+ replaced
WARN()withpr_err()+ renamed
kernel/printk/printk.h->internal.hQuoted 7 lines
Quoted 1 line
+ fixed build on ARM (undefined
NMI_LOG_BUF_SHIFT)So the review of the v3 patchset was ... inconclusive. And everyone has gone quiet about v4.
Probably because you didn't cc the V3 discussion participants when sending out V4. Big mistake, sorry, I can't check everything!
Ah, v4 was sent too fast after v3 before others reacted. It included rather cosmetic changes based on your (Andrew's) feedback.
Quoted 2 lines
But v4 is basically unaltered from v3 so can we please rev this up again? yay or nay? Thanks.
The patch set prevents deadlocks that happen in a real life. It increases a chance to get a valid crash dump when NMI is involved.
Quoted 7 lines
It does not handle well the situation when NMI is involved and the crash dump could not get produced. But this is not handled well even now. As Peter Zijlstra writes this a usually a real mess when he needs to use special hacks (
early_printk) anyway.I still believe that the patch set makes sense and is acceptable as is.
Is anyone against getting this into the mainline for 4.6, please?
Best Regards,
Petr