Florian Weimer writes:
> * Mark Wielaard:
>
>> On Wed, 2023-01-18 at 07:22 +0100, Florian Weimer wrote:
>
>>> If the unwind information is incomplete, this …
>>>
>>> > 7) signal handler unwinds the calling thread however it wants (and can
>>> > sleep and take page faults if needed)
>>>
>>> … m
Mark Wielaard writes:
> Hi Florian,
>
> On Wed, 2023-01-18 at 07:22 +0100, Florian Weimer wrote:
>> * Daniel Colascione:
>>
>> > See, both pro-FP and anti-FP camps think that it's the kernel that has
>> > to do the unwinding unless we copy whole stacks into traces.
>>
>> Well, I think we shoul
Florian Weimer writes:
> * Daniel Colascione:
>
>> See, both pro-FP and anti-FP camps think that it's the kernel that has
>> to do the unwinding unless we copy whole stacks into traces.
>
> Well, I think we should explore hardware-assisted backtraces (shadow
> stacks), which hopefully are going
Hi Florian,
On Wed, 2023-01-18 at 15:01 +0100, Florian Weimer wrote:
> We won't have unwind data for JIT-compiled code, including libffi
> trampolines. We could stop backtracing there (what does the ABI say
> about frames without unwinding information?), but I'm not sure if
> that's
> going to be
On Wed, Jan 18, 2023 at 03:01:19PM +0100, Florian Weimer wrote:
> > On Wed, 2023-01-18 at 07:22 +0100, Florian Weimer wrote:
>
> >> If the unwind information is incomplete, this …
> >>
> >> > 7) signal handler unwinds the calling thread however it wants (and can
> >> > sleep and take page faults
* Mark Wielaard:
> On Wed, 2023-01-18 at 07:22 +0100, Florian Weimer wrote:
>> If the unwind information is incomplete, this …
>>
>> > 7) signal handler unwinds the calling thread however it wants (and can
>> > sleep and take page faults if needed)
>>
>> … might encounter segmentation faults an
Hi Florian,
On Wed, 2023-01-18 at 07:22 +0100, Florian Weimer wrote:
> * Daniel Colascione:
>
> > See, both pro-FP and anti-FP camps think that it's the kernel that has
> > to do the unwinding unless we copy whole stacks into traces.
>
> Well, I think we should explore hardware-assisted backtrac
* Daniel Colascione:
> See, both pro-FP and anti-FP camps think that it's the kernel that has
> to do the unwinding unless we copy whole stacks into traces.
Well, I think we should explore hardware-assisted backtraces (shadow
stacks), which hopefully are going to get merged in Linux 6.2.
> Why s
On 1/17/23 19:21, Mark Wielaard wrote:
> Hi Daniel,
>
> On Mon, Jan 16, 2023 at 08:30:21PM -, Daniel Colascione wrote:
>> As mentioned in [1], instead of finding a way to have the kernel
>> unwind user programs, we can create a protocol through which the
>> kernel can ask usermode to unwind it
Hi Daniel,
On Mon, Jan 16, 2023 at 08:30:21PM -, Daniel Colascione wrote:
> As mentioned in [1], instead of finding a way to have the kernel
> unwind user programs, we can create a protocol through which the
> kernel can ask usermode to unwind itself.
I like this idea and was discussing and t
On Mon, Jan 16, 2023 at 11:33 PM Daniel Colascione wrote:
>
> > Having the vDSO do the unwinding allows the unwinding to be entirely
> transparent to userspace programs and libraries
>
> Why *should* unwinding be transparent to userspace programs and libraries?
> Userspace can contribute to makin
> Having the vDSO do the unwinding allows the unwinding to be entirely
transparent to userspace programs and libraries
Why *should* unwinding be transparent to userspace programs and libraries?
Userspace can contribute to making unwinding better, especially in the managed
case, by hooking into t
On 1/16/23 16:32, Daniel Colascione wrote:
>> Could the vDSO do the unwinding?
>
> The vDSO is just userspace code that happens to be provided by the kernel,
> so, sure, a function in vDSO *could* unwind. But why would it? What would be
> the advantage of doing that over putting the unwinding in
> Could the vDSO do the unwinding?
The vDSO is just userspace code that happens to be provided by the kernel, so,
sure, a function in vDSO *could* unwind. But why would it? What would be the
advantage of doing that over putting the unwinding in libc? To change the vDSO,
you have to change the k
On 1/16/23 15:54, Daniel Colascione wrote:
>> On Mon, Jan 16, 2023 at 3:30 PM Daniel Colascione > wrote:
>>
>> This sounds great, but how is it going to get made?
> Someone has to do it. 😄 I've been thinking about adding this mechanism for a
> few years, but haven't had time so far. I suppose the
> On Mon, Jan 16, 2023 at 3:30 PM Daniel Colascione wrote:
>
> This sounds great, but how is it going to get made?
Someone has to do it. :-) I've been thinking about adding this mechanism for a
few years, but haven't had time so far. I suppose the first step would be
raising the subject on li
On Mon, Jan 16, 2023 at 3:30 PM Daniel Colascione wrote:
>
> Frame pointers also have the disadvantage of working only with AOT-compiled
> languages for which a trace analysis tool can associate an instruction
> pointer with a semantically-relevant bit of code. If you try to use frame
> pointer
17 matches
Mail list logo
