Skip to contentSkip to conversation

GDB: Support getting current task struct in UML

PATCH Superseded Glenn Washburn 5 participants 13 messages Feb 16, 2023 – Feb 26, 2023

Subsystems (2)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

v2 cover GDB: Support getting current task struct in UML

Added suggestions from Jan.

Glenn

Glenn Washburn (2):
scripts/gdb: Correct indentation in get_current_task scripts/gdb: Support getting current task struct in UML

 scripts/gdb/linux/cpus.py | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

Range-diff against v1:
1: f33ebe524590 ! 1: c5a916e094d9 scripts/gdb: Correct indentation in get_current_task
@@ Commit message

There is an extra space in a couple blocks in get_current_task. Though python does not care, let's make the spacing consistent.
+ Also, format better an if expression, removing unneeded parenthesis.

## scripts/gdb/linux/cpus.py ##
@@ scripts/gdb/linux/cpus.py: def get_current_task(cpu):
@@ scripts/gdb/linux/cpus.py: def get_current_task(cpu):
- raise gdb.GdbError("Sorry, obtaining the current task is not allowed "
- "while running in userspace(EL0)")
+ current_task_addr = gdb.parse_and_eval("$SP_EL0")
-+ if((current_task_addr >> 63) != 0):
++ if (current_task_addr >> 63) != 0:
+ current_task = current_task_addr.cast(task_ptr_type)
+ return current_task.dereference()
+ else:
2: 8c3db2291f58 ! 2: 683d10e752cd scripts/gdb: Support getting current task struct in UML
@@ Commit message
A running x86 UML kernel reports with architecture "i386:x86-64" as it is a sub-architecture. However, a difference with bare-metal x86 kernels is in how it manages tasks and the current task struct. To
- identify that the inferior is a UML kernel and not bare-metal, the
- symbol "uml_kmalloc" is checked for. If it exists, then do the UML
- specific way of getting the current task struct.
+ identify that the inferior is a UML kernel and not bare-metal, check
+ for the existence of the UML specific symbol "cpu_tasks" which
+ contains the current task struct.

## scripts/gdb/linux/cpus.py ##
@@ scripts/gdb/linux/cpus.py: def get_current_task(cpu):
@@ scripts/gdb/linux/cpus.py: def get_current_task(cpu):
if utils.is_target_arch("x86"):
- var_ptr = gdb.parse_and_eval("&current_task")
- return per_cpu(var_ptr, cpu).dereference()
-+ if gdb.lookup_global_symbol("uml_kmalloc"):
++ if gdb.lookup_global_symbol("cpu_tasks"):
++ # This is a UML kernel, which stores the current task
++ # differently than other x86 sub architectures
+ var_ptr = gdb.parse_and_eval("(struct task_struct *)cpu_tasks[0].task")
+ return var_ptr.dereference()
+ else:
@@ scripts/gdb/linux/cpus.py: def get_current_task(cpu):
+ return per_cpu(var_ptr, cpu).dereference()
elif utils.is_target_arch("aarch64"):
current_task_addr = gdb.parse_and_eval("$SP_EL0")
- if((current_task_addr >> 63) != 0):
+ if (current_task_addr >> 63) != 0:

Signature · 7 lines
2.30.2


_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um
#1 lore raw

PATCH v2 1/2 scripts/gdb: Correct indentation in get_current_task

There is an extra space in a couple blocks in get_current_task. Though python does not care, let's make the spacing consistent. Also, format better an if expression, removing unneeded parenthesis.

---
 scripts/gdb/linux/cpus.py | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)
scripts/gdb/linux/cpus.py +9 -9
index 15fc4626d236..3e02a1866751 100644
@@ -163,16 +163,16 @@ def get_current_task(cpu):
     task_ptr_type = task_type.get_type().pointer()
 
     if utils.is_target_arch("x86"):
-         var_ptr = gdb.parse_and_eval("&current_task")
-         return per_cpu(var_ptr, cpu).dereference()
+        var_ptr = gdb.parse_and_eval("&current_task")
+        return per_cpu(var_ptr, cpu).dereference()
     elif utils.is_target_arch("aarch64"):
-         current_task_addr = gdb.parse_and_eval("$SP_EL0")
-         if((current_task_addr >> 63) != 0):
-             current_task = current_task_addr.cast(task_ptr_type)
-             return current_task.dereference()
-         else:
-             raise gdb.GdbError("Sorry, obtaining the current task is not allowed "
-                                "while running in userspace(EL0)")
+        current_task_addr = gdb.parse_and_eval("$SP_EL0")
+        if (current_task_addr >> 63) != 0:
+            current_task = current_task_addr.cast(task_ptr_type)
+            return current_task.dereference()
+        else:
+            raise gdb.GdbError("Sorry, obtaining the current task is not allowed "
+                               "while running in userspace(EL0)")
     else:
         raise gdb.GdbError("Sorry, obtaining the current task is not yet "
                            "supported with this arch")
Signature · 7 lines
2.30.2


_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um
#2 lore raw

PATCH v2 2/2 scripts/gdb: Support getting current task struct in UML

In reply to Glenn Washburn

A running x86 UML kernel reports with architecture "i386:x86-64" as it is a sub-architecture. However, a difference with bare-metal x86 kernels is in how it manages tasks and the current task struct. To identify that the inferior is a UML kernel and not bare-metal, check for the existence of the UML specific symbol "cpu_tasks" which contains the current task struct.

---
 scripts/gdb/linux/cpus.py | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
scripts/gdb/linux/cpus.py +8 -2
index 3e02a1866751..4b4ce6464dee 100644
@@ -163,8 +163,14 @@ def get_current_task(cpu):
     task_ptr_type = task_type.get_type().pointer()
 
     if utils.is_target_arch("x86"):
-        var_ptr = gdb.parse_and_eval("&current_task")
-        return per_cpu(var_ptr, cpu).dereference()
+        if gdb.lookup_global_symbol("cpu_tasks"):
+            # This is a UML kernel, which stores the current task
+            # differently than other x86 sub architectures
+            var_ptr = gdb.parse_and_eval("(struct task_struct *)cpu_tasks[0].task")
+            return var_ptr.dereference()
+        else:
+            var_ptr = gdb.parse_and_eval("&current_task")
+            return per_cpu(var_ptr, cpu).dereference()
     elif utils.is_target_arch("aarch64"):
         current_task_addr = gdb.parse_and_eval("$SP_EL0")
         if (current_task_addr >> 63) != 0:
Signature · 7 lines
2.30.2


_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um
maintainer Reviewed-by on the v2 series
#3 lore raw
In reply to Glenn Washburn
Quoted 53 lines · Glenn Washburn

Added suggestions from Jan.

Glenn

Glenn Washburn (2):
scripts/gdb: Correct indentation in get_current_task scripts/gdb: Support getting current task struct in UML

 scripts/gdb/linux/cpus.py | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

Range-diff against v1:
1: f33ebe524590 ! 1: c5a916e094d9 scripts/gdb: Correct indentation in get_current_task
@@ Commit message

There is an extra space in a couple blocks in get_current_task. Though python does not care, let's make the spacing consistent.
+ Also, format better an if expression, removing unneeded parenthesis.

## scripts/gdb/linux/cpus.py ##
@@ scripts/gdb/linux/cpus.py: def get_current_task(cpu):
@@ scripts/gdb/linux/cpus.py: def get_current_task(cpu):
- raise gdb.GdbError("Sorry, obtaining the current task is not allowed "
- "while running in userspace(EL0)")
+ current_task_addr = gdb.parse_and_eval("$SP_EL0")
-+ if((current_task_addr >> 63) != 0):
++ if (current_task_addr >> 63) != 0:
+ current_task = current_task_addr.cast(task_ptr_type)
+ return current_task.dereference()
+ else:
2: 8c3db2291f58 ! 2: 683d10e752cd scripts/gdb: Support getting current task struct in UML
@@ Commit message
A running x86 UML kernel reports with architecture "i386:x86-64" as it is a sub-architecture. However, a difference with bare-metal x86 kernels is in how it manages tasks and the current task struct. To
- identify that the inferior is a UML kernel and not bare-metal, the
- symbol "uml_kmalloc" is checked for. If it exists, then do the UML
- specific way of getting the current task struct.
+ identify that the inferior is a UML kernel and not bare-metal, check
+ for the existence of the UML specific symbol "cpu_tasks" which
+ contains the current task struct.

## scripts/gdb/linux/cpus.py ##
@@ scripts/gdb/linux/cpus.py: def get_current_task(cpu):
@@ scripts/gdb/linux/cpus.py: def get_current_task(cpu):
if utils.is_target_arch("x86"):
- var_ptr = gdb.parse_and_eval("&current_task")
- return per_cpu(var_ptr, cpu).dereference()
-+ if gdb.lookup_global_symbol("uml_kmalloc"):
++ if gdb.lookup_global_symbol("cpu_tasks"):
++ # This is a UML kernel, which stores the current task
++ # differently than other x86 sub architectures
+ var_ptr = gdb.parse_and_eval("(struct task_struct *)cpu_tasks[0].task")

Quoted 8 lines

+ return var_ptr.dereference()
+ else:
@@ scripts/gdb/linux/cpus.py: def get_current_task(cpu):
+ return per_cpu(var_ptr, cpu).dereference()
elif utils.is_target_arch("aarch64"):
current_task_addr = gdb.parse_and_eval("$SP_EL0")
- if((current_task_addr >> 63) != 0):
+ if (current_task_addr >> 63) != 0:

Jan

Signature · 8 lines
Siemens AG, Technology
Competence Center Embedded Linux


_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um
#4 lore raw
Quoted 5 lines · On Thu, 16 Feb 2023 08:51:52 +0100 Jan Kiszka <[email protected]> wrote:
Quoted 4 lines · Glenn Washburn

Added suggestions from Jan.

Glenn

Quoted 9 lines
Quoted 4 lines

Glenn Washburn (2):
scripts/gdb: Correct indentation in get_current_task scripts/gdb: Support getting current task struct in UML

...

Thanks, but I'm not subscribed to linux-um and my usual get-it-from-lkml didn't work.

Could we please have a resend, with a cc to linux-kernel?

_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um

#5 lore raw

----- Ursprüngliche Mail -----

Quoted 5 lines

Von: "Andrew Morton" <[email protected]> Thanks, but I'm not subscribed to linux-um and my usual get-it-from-lkml didn't work.

Could we please have a resend, with a cc to linux-kernel?

You can also get the patches in mbox format from:
https://patchwork.ozlabs.org/project/linux-um/list/?series=342212

Thanks,
//richard

_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um

#6 lore raw
Quoted 9 lines · On Thu, 16 Feb 2023 23:54:38 +0100 (CET) Richard Weinberger <[email protected]>…

----- Ursprüngliche Mail -----

Quoted 5 lines

Von: "Andrew Morton" <[email protected]> Thanks, but I'm not subscribed to linux-um and my usual get-it-from-lkml didn't work.

Could we please have a resend, with a cc to linux-kernel?

You can also get the patches in mbox format from:
https://patchwork.ozlabs.org/project/linux-um/list/?series=342212

Thanks, but..

- Can't do reply-to-all if I have comments
- Can't collect people's followup acks and review comments
- No Link tag

etc.

_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um

#7 lore raw

Hi Andrew,

Quoted 10 lines · Andrew Morton
Quoted 9 lines · On Thu, 16 Feb 2023 23:54:38 +0100 (CET) Richard Weinberger <[email protected]>…

----- Ursprüngliche Mail -----

Quoted 5 lines

Von: "Andrew Morton" <[email protected]> Thanks, but I'm not subscribed to linux-um and my usual get-it-from-lkml didn't work.

Could we please have a resend, with a cc to linux-kernel?

You can also get the patches in mbox format from:
https://patchwork.ozlabs.org/project/linux-um/list/?series=342212

And from there, you can get the Message-IDs, and use lore:
https://lore.kernel.org/r/683d10e752cd4852ac62ef3cc3e9a6972a017bdf.1676532759.git.development@efficientek.com
or b4:
b4 mbox 683d10e752cd4852ac62ef3cc3e9a6972a017bdf.1676532759.git.development@efficientek.com b4 am 683d10e752cd4852ac62ef3cc3e9a6972a017bdf.1676532759.git.development@efficientek.com

Quoted 5 lines

Thanks, but..

- Can't do reply-to-all if I have comments
- Can't collect people's followup acks and review comments
- No Link tag

Gr{oetje,eeting}s,

Geert

Signature · 10 lines
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um
#8 lore raw
Quoted 16 lines · On Fri, 17 Feb 2023 11:14:01 +0100 Geert Uytterhoeven <[email protected]>…

Hi Andrew,

Quoted 11 lines · Andrew Morton
Quoted 9 lines · On Thu, 16 Feb 2023 23:54:38 +0100 (CET) Richard Weinberger <[email protected]>…

----- Ursprüngliche Mail -----

Quoted 5 lines

Von: "Andrew Morton" <[email protected]> Thanks, but I'm not subscribed to linux-um and my usual get-it-from-lkml didn't work.

Could we please have a resend, with a cc to linux-kernel?

You can also get the patches in mbox format from:
https://patchwork.ozlabs.org/project/linux-um/list/?series=342212

Quoted 13 lines

And from there, you can get the Message-IDs, and use lore:
https://lore.kernel.org/r/683d10e752cd4852ac62ef3cc3e9a6972a017bdf.1676532759.git.development@efficientek.com
or b4:
b4 mbox 683d10e752cd4852ac62ef3cc3e9a6972a017bdf.1676532759.git.development@efficientek.com b4 am 683d10e752cd4852ac62ef3cc3e9a6972a017bdf.1676532759.git.development@efficientek.com

Quoted 5 lines

Thanks, but..

- Can't do reply-to-all if I have comments
- Can't collect people's followup acks and review comments
- No Link tag

Ok, so do I need to resend after all? And anther point in need of clarification, regardless of subsystem or area of source, when sending patches, should I always cc to linux-kernel?

Glenn

_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um

#9 lore raw
Quoted 27 lines · On Fri, 17 Feb 2023 15:56:43 -0600 Glenn Washburn…
Quoted 24 lines · On Fri, 17 Feb 2023 11:14:01 +0100 Geert Uytterhoeven <[email protected]>…

Hi Andrew,

Quoted 11 lines · Andrew Morton
Quoted 9 lines · On Thu, 16 Feb 2023 23:54:38 +0100 (CET) Richard Weinberger <[email protected]>…

----- Ursprüngliche Mail -----

Quoted 5 lines

Von: "Andrew Morton" <[email protected]> Thanks, but I'm not subscribed to linux-um and my usual get-it-from-lkml didn't work.

Could we please have a resend, with a cc to linux-kernel?

You can also get the patches in mbox format from:
https://patchwork.ozlabs.org/project/linux-um/list/?series=342212

And from there, you can get the Message-IDs, and use lore:
https://lore.kernel.org/r/683d10e752cd4852ac62ef3cc3e9a6972a017bdf.1676532759.git.development@efficientek.com
or b4:
b4 mbox 683d10e752cd4852ac62ef3cc3e9a6972a017bdf.1676532759.git.development@efficientek.com b4 am 683d10e752cd4852ac62ef3cc3e9a6972a017bdf.1676532759.git.development@efficientek.com

Quoted 7 lines
Quoted 5 lines
Quoted 5 lines

Thanks, but..

- Can't do reply-to-all if I have comments
- Can't collect people's followup acks and review comments
- No Link tag

Ok, so do I need to resend after all?

Yes please. Call it v2, add any acks and reviewed-bys.

Quoted 3 lines

And anther point in need of
clarification, regardless of subsystem or area of source, when sending patches, should I always cc to linux-kernel?

I think it's best. That's pretty much the only purpose lkml serves now :(

_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um

#10 lore raw
In reply to Glenn Washburn
Quoted 8 lines · On Fri, 17 Feb 2023 19:11:52 -0600 Glenn Washburn…

Added suggestions from Jan.

Glenn

Glenn Washburn (2):
scripts/gdb: Correct indentation in get_current_task scripts/gdb: Support getting current task struct in UML

For some reason I get a bunch of rejects when applying these on top of the latest patchpile. Please check my end result:

def get_current_task(cpu):
task_ptr_type = task_type.get_type().pointer()

if utils.is_target_arch("x86"):
if gdb.lookup_global_symbol("cpu_tasks"):
# This is a UML kernel, which stores the current task
# differently than other x86 sub architectures
var_ptr = gdb.parse_and_eval("(struct task_struct *)cpu_tasks[0].task") return var_ptr.dereference()
else:
var_ptr = gdb.parse_and_eval("&current_task")
return per_cpu(var_ptr, cpu).dereference()
elif utils.is_target_arch("aarch64"):
current_task_addr = gdb.parse_and_eval("$SP_EL0") if((current_task_addr >> 63) != 0):
current_task = current_task_addr.cast(task_ptr_type) return current_task.dereference()
else:
raise gdb.GdbError("Sorry, obtaining the current task is not allowed "
"while running in userspace(EL0)")
else:
raise gdb.GdbError("Sorry, obtaining the current task is not yet "
"supported with this arch")

#11 lore raw
Quoted 7 lines · On Thu, 23 Feb 2023 13:14:02 -0800 Andrew Morton <[email protected]>…
Quoted 4 lines · On Fri, 17 Feb 2023 19:11:52 -0600 Glenn Washburn…

Added suggestions from Jan.

Glenn

Quoted 7 lines
Quoted 4 lines

Glenn Washburn (2):
scripts/gdb: Correct indentation in get_current_task scripts/gdb: Support getting current task struct in UML

For some reason I get a bunch of rejects when applying these on top of the latest patchpile. Please check my end result:

It looks like there was a change from v6.2-rc8 to v6.2 that caused a conflict.

Quoted 10 lines

def get_current_task(cpu):
task_ptr_type = task_type.get_type().pointer()

if utils.is_target_arch("x86"):
if gdb.lookup_global_symbol("cpu_tasks"):
# This is a UML kernel, which stores the current task # differently than other x86 sub architectures var_ptr = gdb.parse_and_eval("(struct task_struct
*)cpu_tasks[0].task") return var_ptr.dereference()

This is missing the return statement in the second patch.

Quoted 2 lines

else:
var_ptr = gdb.parse_and_eval("&current_task")

It looks like "current_task" has now been changed to "pcpu_hot.current_task" in v6.2.

Would you like me to resent the series rebased onto v6.2?

Glenn

Quoted 15 lines

return per_cpu(var_ptr, cpu).dereference()
elif utils.is_target_arch("aarch64"):
current_task_addr = gdb.parse_and_eval("$SP_EL0") if((current_task_addr >> 63) != 0):
current_task = current_task_addr.cast(task_ptr_type) return current_task.dereference()
else:
raise gdb.GdbError("Sorry, obtaining the current task is
not allowed " "while running in userspace(EL0)")
else:
raise gdb.GdbError("Sorry, obtaining the current task is not
yet " "supported with this arch")

#12 lore raw
Quoted 1 line · On Sun, 26 Feb 2023 00:47:19 -0600 Glenn Washburn…

Would you like me to resent the series rebased onto v6.2?

Yes please ;) Against current linus tip if possible.