Changes in kgdb.c

e6ce4a48 - MIPS: kernel: Drop kgdb_call_nmi_hook

Sat, 13 Feb 2021 14:02:31 +0000

MIPS: kernel: Drop kgdb_call_nmi_hookWith the removal of set_fs() calls kgdb_call_nmi_hook() is now the same asthe default implementation, so we can remove it.Signed-off-by: Thomas Bogendoerfer List of files: /linux-6.15/arch/mips/kernel/kgdb.c

b1468f30 - Revert "MIPS: kernel: {ftrace,kgdb}: Set correct address limit for cache flushes"

Wed, 10 Feb 2021 16:16:13 +0000

Revert "MIPS: kernel: {ftrace,kgdb}: Set correct address limit for cache flushes"This reverts commit 6ebda44f366478d1eea180d93154e7d97b591f50.All icache flushes in this code paths are done via flush_icache_range(),which only uses normal cache instruction. And this is the correct thingfor EVA mode, too. So no need to do set_fs(KERNEL_DS) here.Signed-off-by: Thomas Bogendoerfer Reviewed-by: Christoph Hellwig List of files: /linux-6.15/arch/mips/kernel/kgdb.c

ab8a6d82 - MIPS: KGDB: fix kgdb support for SMP platforms.

Wed, 27 Mar 2019 23:08:01 +0000

MIPS: KGDB: fix kgdb support for SMP platforms.KGDB_call_nmi_hook is called by other cpu through smp call.MIPS smp call is processed in ipi irq handler and regs is saved in handle_int.So kgdb_call_nmi_hook get regs by get_irq_regs and regs will be passed to kgdb_cpu_enter.Signed-off-by: Chong Qiao Reviewed-by: Douglas Anderson Acked-by: Daniel Thompson Signed-off-by: Paul Burton Cc: Ralf Baechle Cc: James Hogan Cc: Will Deacon Cc: Christophe Leroy Cc: linux-mips@vger.kernel.orgCc: linux-kernel@vger.kernel.orgCc: QiaoChong List of files: /linux-6.15/arch/mips/kernel/kgdb.c

736706be - get rid of legacy 'get_ds()' function

Mon, 04 Mar 2019 18:39:05 +0000

get rid of legacy 'get_ds()' functionEvery in-kernel use of this function defined it to KERNEL_DS (either asan actual define, or as an inline function). It's an entirelyhistorical artifact, and long long long ago used to actually read thesegment selector valueof '%ds' on x86.Which in the kernel is always KERNEL_DS.Inspired by a patch from Jann Horn that just did this for a very smallsubset of users (the ones in fs/), along with Al who suggested a script.I then just took it to the logical extreme and removed all the remaininggunk.Roughly scripted with git grep -l '(get_ds())' -- :^tools/ | xargs sed -i 's/(get_ds())/(KERNEL_DS)/' git grep -lw 'get_ds' -- :^tools/ | xargs sed -i '/^#define get_ds()/d'plus manual fixups to remove a few unusual usage patterns, the couple ofinline function cases and to fix up a comment that had become stale.The 'get_ds()' function remains in an x86 kvm selftest, since in userspace it actually does something relevant.Inspired-by: Jann Horn Inspired-by: Al Viro Signed-off-by: Linus Torvalds List of files: /linux-6.15/arch/mips/kernel/kgdb.c

cc028297 - kgdb/treewide: constify struct kgdb_arch arch_kgdb_ops

Thu, 06 Dec 2018 20:07:40 +0000

kgdb/treewide: constify struct kgdb_arch arch_kgdb_opscheckpatch.pl reports the following: WARNING: struct kgdb_arch should normally be const #28: FILE: arch/mips/kernel/kgdb.c:397: +struct kgdb_arch arch_kgdb_ops = {This report makes sense, as all other ops struct, thisone should also be const. This patch does the change.Cc: Vineet Gupta Cc: Russell King Cc: Catalin Marinas Cc: Will Deacon Cc: Yoshinori Sato Cc: Richard Kuo Cc: Michal Simek Cc: Ralf Baechle Cc: Paul Burton Cc: James Hogan Cc: Ley Foon Tan Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Rich Felker Cc: "David S. Miller" Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: x86@kernel.orgAcked-by: Daniel Thompson Acked-by: Paul Burton Signed-off-by: Christophe Leroy Acked-by: Borislav Petkov Acked-by: Michael Ellerman (powerpc)Signed-off-by: Daniel Thompson List of files: /linux-6.15/arch/mips/kernel/kgdb.c

911b7afd - mips/kgdb: prepare arch_kgdb_ops for constness

Thu, 06 Dec 2018 20:07:38 +0000

mips/kgdb: prepare arch_kgdb_ops for constnessMIPS is the only architecture modifying arch_kgdb_ops during init.This patch makes the init static, so that it can be changed toconst in following patch, as recommended by checkpatch.plSuggested-by: Paul Burton Acked-by: Daniel Thompson Acked-by: Paul Burton Signed-off-by: Christophe Leroy Signed-off-by: Daniel Thompson List of files: /linux-6.15/arch/mips/kernel/kgdb.c

3cd99ac3 - kgdb: Fix kgdb_roundup_cpus() for arches who used smp_call_function()

Wed, 05 Dec 2018 03:38:26 +0000

kgdb: Fix kgdb_roundup_cpus() for arches who used smp_call_function()When I had lockdep turned on and dropped into kgdb I got a nice splaton my system. Specifically it hit: DEBUG_LOCKS_WARN_ON(current->hardirq_context)Specifically it looked like this: sysrq: SysRq : DEBUG ------------[ cut here ]------------ DEBUG_LOCKS_WARN_ON(current->hardirq_context) WARNING: CPU: 0 PID: 0 at .../kernel/locking/lockdep.c:2875 lockdep_hardirqs_on+0xf0/0x160 CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.19.0 #27 pstate: 604003c9 (nZCv DAIF +PAN -UAO) pc : lockdep_hardirqs_on+0xf0/0x160 ... Call trace: lockdep_hardirqs_on+0xf0/0x160 trace_hardirqs_on+0x188/0x1ac kgdb_roundup_cpus+0x14/0x3c kgdb_cpu_enter+0x53c/0x5cc kgdb_handle_exception+0x180/0x1d4 kgdb_compiled_brk_fn+0x30/0x3c brk_handler+0x134/0x178 do_debug_exception+0xfc/0x178 el1_dbg+0x18/0x78 kgdb_breakpoint+0x34/0x58 sysrq_handle_dbg+0x54/0x5c __handle_sysrq+0x114/0x21c handle_sysrq+0x30/0x3c qcom_geni_serial_isr+0x2dc/0x30c ... ... irq event stamp: ...45 hardirqs last enabled at (...44): [...] __do_softirq+0xd8/0x4e4 hardirqs last disabled at (...45): [...] el1_irq+0x74/0x130 softirqs last enabled at (...42): [...] _local_bh_enable+0x2c/0x34 softirqs last disabled at (...43): [...] irq_exit+0xa8/0x100 ---[ end trace adf21f830c46e638 ]---Looking closely at it, it seems like a really bad idea to be callinglocal_irq_enable() in kgdb_roundup_cpus(). If nothing else that seemslike it could violate spinlock semantics and cause a deadlock.Instead, let's use a private csd alongsidesmp_call_function_single_async() to round up the other CPUs. Usingsmp_call_function_single_async() doesn't require interrupts to beenabled so we can remove the offending bit of code.In order to avoid duplicating this across all the architectures thatuse the default kgdb_roundup_cpus(), we'll add a "weak" implementationto debug_core.c.Looking at all the people who previously had copies of this code,there were a few variants. I've attempted to keep the variantsworking like they used to. Specifically:* For arch/arc we passed NULL to kgdb_nmicallback() instead of get_irq_regs().* For arch/mips there was a bit of extra code around kgdb_nmicallback()NOTE: In this patch we will still get into trouble if we try to roundup a CPU that failed to round up before. We'll try to round it upagain and potentially hang when we try to grab the csd lock. That'snot new behavior but we'll still try to do better in a future patch.Suggested-by: Daniel Thompson Signed-off-by: Douglas Anderson Cc: Vineet Gupta Cc: Russell King Cc: Catalin Marinas Cc: Will Deacon Cc: Richard Kuo Cc: Ralf Baechle Cc: Paul Burton Cc: James Hogan Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Yoshinori Sato Cc: Rich Felker Cc: "David S. Miller" Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: "H. Peter Anvin" Acked-by: Will Deacon Signed-off-by: Daniel Thompson List of files: /linux-6.15/arch/mips/kernel/kgdb.c

9ef7fa50 - kgdb: Remove irq flags from roundup

Wed, 05 Dec 2018 03:38:25 +0000

kgdb: Remove irq flags from roundupThe function kgdb_roundup_cpus() was passed a parameter that wasdocumented as:> the flags that will be used when restoring the interrupts. There is> local_irq_save() call before kgdb_roundup_cpus().Nobody used those flags. Anyone who wanted to temporarily turn oninterrupts just did local_irq_enable() and local_irq_disable() withoutlooking at them. So we can definitely remove the flags.Signed-off-by: Douglas Anderson Cc: Vineet Gupta Cc: Russell King Cc: Catalin Marinas Cc: Will Deacon Cc: Richard Kuo Cc: Ralf Baechle Cc: Paul Burton Cc: James Hogan Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Yoshinori Sato Cc: Rich Felker Cc: "David S. Miller" Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: "H. Peter Anvin" Acked-by: Will Deacon Signed-off-by: Daniel Thompson List of files: /linux-6.15/arch/mips/kernel/kgdb.c

162b270c - MIPS: KGDB: Use kernel context for sleeping threads

Thu, 30 Mar 2017 15:06:02 +0000

MIPS: KGDB: Use kernel context for sleeping threadsKGDB is a kernel debug stub and it can't be used to debug userland as itcan only safely access kernel memory.On MIPS however KGDB has always got the register state of sleepingprocesses from the userland register context at the beginning of thekernel stack. This is meaningless for kernel threads (which never enteruserland), and for user threads it prevents the user seeing what it isdoing while in the kernel:(gdb) info threads Id Target Id Frame ... 3 Thread 2 (kthreadd) 0x0000000000000000 in ?? () 2 Thread 1 (init) 0x000000007705c4b4 in ?? () 1 Thread -2 (shadowCPU0) 0xffffffff8012524c in arch_kgdb_breakpoint () at arch/mips/kernel/kgdb.c:201Get the register state instead from the (partial) kernel registercontext stored in the task's thread_struct for resume() to restore. Allthreads now correctly appear to be in context_switch():(gdb) info threads Id Target Id Frame ... 3 Thread 2 (kthreadd) context_switch (rq=, cookie=..., next=, prev=0x0) at kernel/sched/core.c:2903 2 Thread 1 (init) context_switch (rq=, cookie=..., next=, prev=0x0) at kernel/sched/core.c:2903 1 Thread -2 (shadowCPU0) 0xffffffff8012524c in arch_kgdb_breakpoint () at arch/mips/kernel/kgdb.c:201Call clobbered registers which aren't saved and exception registers(BadVAddr & Cause) which can't be easily determined without stackunwinding are reported as 0. The PC is taken from the return address,such that the state presented matches that found immediately afterreturning from resume().Fixes: 8854700115ec ("[MIPS] kgdb: add arch support for the kernel's kgdb core")Signed-off-by: James Hogan Cc: Jason Wessel Cc: linux-mips@linux-mips.orgCc: stable@vger.kernel.orgPatchwork: https://patchwork.linux-mips.org/patch/15829/Signed-off-by: Ralf Baechle List of files: /linux-6.15/arch/mips/kernel/kgdb.c

7c0f6ba6 - Replace with globally

Sat, 24 Dec 2016 19:46:01 +0000

Replace with globallyThis was entirely automated, using the script by Al: PATT='^[[:blank:]]*#[[:blank:]]*include[[:blank:]]*' sed -i -e "s!$PATT!#include !" \ $(git grep -l "$PATT"|grep -v ^include/linux/uaccess.h)to do the replacement at the end of the merge window.Requested-by: Al Viro Signed-off-by: Linus Torvalds List of files: /linux-6.15/arch/mips/kernel/kgdb.c

52ea7bff - MIPS: Get rid of 'kgdb_early_setup' cruft.

Sun, 31 May 2015 21:40:32 +0000

MIPS: Get rid of 'kgdb_early_setup' cruft.Commit 854700115ecf ([MIPS] kgdb: add arch support for the kernel's kgdb core)added the 'kgdb_early_setup' flag to avoid calling trap_init() and init_IRQ()the second time, however the code that called these functions earlier, fromkgdb_arch_init(), had been already removed by that time, so the flag neverserved any useful purpose. Remove the related code along with ugly #ifdef'eryat last.[ralf@linux-mips.org: Folded in Guenter Roeck's fix.]Signed-off-by: Sergei Shtylyov Patchwork: https://patchwork.linux-mips.org/patch/10501/Signed-off-by: Guenter Roeck Patchwork: https://patchwork.linux-mips.org/patch/10533/Signed-off-by: Ralf Baechle List of files: /linux-6.15/arch/mips/kernel/kgdb.c

6ebda44f - MIPS: kernel: {ftrace,kgdb}: Set correct address limit for cache flushes

Mon, 16 Dec 2013 12:06:55 +0000

MIPS: kernel: {ftrace,kgdb}: Set correct address limit for cache flushesWhen flushing the icache, make sure the address limit is correctso the appropriate 'cache' instruction will be used. This has noimpact on cores operating in non-eva mode. However, when EVA isenabled, we ensure that 'cache' will be used instead of 'cachee'.Signed-off-by: Leonid Yegoshin Signed-off-by: Markos Chandras List of files: /linux-6.15/arch/mips/kernel/kgdb.c

70342287 - MIPS: Whitespace cleanup.

Tue, 22 Jan 2013 11:59:30 +0000

MIPS: Whitespace cleanup.Having received another series of whitespace patches I decided to do thisonce and for all rather than dealing with this kind of patches tricklingin forever.Signed-off-by: Ralf Baechle List of files: /linux-6.15/arch/mips/kernel/kgdb.c

f0a996ee - mips,kgdb: fix recursive page fault with CONFIG_KPROBES

Fri, 10 Aug 2012 17:21:15 +0000

mips,kgdb: fix recursive page fault with CONFIG_KPROBESThis fault was detected using the kgdb test suite on boot and itcrashes recursively due to the fact that CONFIG_KPROBES on mips addsan extra die notifier in the page fault handler. The crash signaturelooks like this:kgdbts:RUN bad memory access testKGDB: re-enter exception: ALL breakpoints killedCall Trace:[<807b7548>] dump_stack+0x20/0x54[<807b7548>] dump_stack+0x20/0x54The fix for now is to have kgdb return immediately if the fault typeis DIE_PAGE_FAULT and allow the kprobe code to decide what is supposedto happen.Cc: Masami Hiramatsu Cc: David S. Miller Cc: Signed-off-by: Jason Wessel List of files: /linux-6.15/arch/mips/kernel/kgdb.c

eefc3f32 - MIPS: Fix a typo.

Thu, 05 Aug 2010 13:51:25 +0000

MIPS: Fix a typo."Userpace" -> "Userspace"Signed-off-by: Andrea Gelmini Cc: Andrea Gelmini Cc: Jason Wessel Cc: Martin Hicks Cc: linux-mips@linux-mips.orgPatchwork: https://patchwork.linux-mips.org/patch/1536/Signed-off-by: Ralf Baechle List of files: /linux-6.15/arch/mips/kernel/kgdb.c

7aeb48b7 - kgdb,mips: remove unused kgdb_cpu_doing_single_step operations

Thu, 05 Aug 2010 14:22:25 +0000

kgdb,mips: remove unused kgdb_cpu_doing_single_step operationsThe mips kgdb specific code does not support software or HW singlestepping so it should not implementSigned-off-by: Jason Wessel Signed-off-by: Dongdong Deng Acked-by: Ralf Baechle CC: linux-mips@linux-mips.org List of files: /linux-6.15/arch/mips/kernel/kgdb.c

0896a9be - kgdb,mips: Individual register get/set for mips

Thu, 05 Aug 2010 14:22:21 +0000

kgdb,mips: Individual register get/set for mipsImplement the ability to individually get and set registers for kdband kgdb for mips.Signed-off-by: Jason Wessel Acked-by: Ralf Baechle CC: linux-mips@linux-mips.org List of files: /linux-6.15/arch/mips/kernel/kgdb.c

5dd11d5d - mips,kgdb: kdb low level trap catch and stack trace

Fri, 21 May 2010 02:04:26 +0000

mips,kgdb: kdb low level trap catch and stack traceThe only way the debugger can handle a trap in inside rcu_lock,notify_die, or atomic_notifier_call_chain without a recursive fault isto have a low level "first opportunity handler" do_trap_or_bp() handler.Generally this will be something the vast majority of folks will notneed, but for those who need it, it is added as a kernel .configoption called KGDB_LOW_LEVEL_TRAP.Also added was a die notification for oops such that kdb can catch anoops for analysis.There appeared to be no obvious way to pass the struct pt_regs fromthe original exception back to the stack back tracer, so a specialcase was added to show_stack() for when kdb is active because yougenerally desire to generally look at the back trace of the originalexception.Signed-off-by: Jason Wessel Acked-by: Ralf Baechle List of files: /linux-6.15/arch/mips/kernel/kgdb.c

dcc78711 - kgdb: core changes to support kdb

Fri, 21 May 2010 02:04:21 +0000

kgdb: core changes to support kdbThese are the minimum changes to the kgdb core in order to enable anAPI to connect a new front end (kdb) to the debug core.This patch introduces the dbg_kdb_mode variable controls where theuser level I/O is routed. It will be routed to the gdbstub (kgdb) orto the kdb front end which is a simple shell available over the kgdbocconnection.You can switch back and forth between kdb or the gdb stub mode ofoperation dynamically. From gdb stub mode you can blindly type"$3#33", or from the kdb mode you can enter "kgdb" to switch to thegdb stub.The logic in the debug core depends on kdb to look for the typical gdbconnection sequences and return immediately with KGDB_PASS_EVENT if agdb serial command sequence is detected. That should allow areasonably seamless transition between kdb -> gdb without leaving thekernel exception state. The two gdb serial queries that kdb isresponsible for detecting are the "?" and "qSupported" packets.CC: Ingo Molnar Signed-off-by: Jason Wessel Acked-by: Martin Hicks List of files: /linux-6.15/arch/mips/kernel/kgdb.c

631330f5 - MIPS: Build fix - include into all smp_processor_id() users.

Fri, 19 Jun 2009 13:05:26 +0000

MIPS: Build fix - include into all smp_processor_id() users.Some of the were relying into smp.h being dragged in by another headerwhich of course is fragile. uses smp_processor_id()only in macros and including smp.h there leads to an include loop, sodon't change cpu-info.h.Signed-off-by: Ralf Baechle List of files: /linux-6.15/arch/mips/kernel/kgdb.c