|
Updated Event: ELISA Safety-Architecture Weekly Meeting - Tuesday, March 22, 2022
#cal-invite
safety-architecture@lists.elisa.tech Calendar <noreply@...>
ELISA Safety-Architecture Weekly Meeting When: Where: Organizer: myu@... myu@... Description:
|
|
|
|
US time shift issue
Hi All Unfortunately for the Arch WG meeting my calendar did not adjust according to the US time shift and now I got a conflict from 1pm to 2pm CET that I cannot resolve. I propose to schedule an extra session next Tue at the usual time (2pm-3pm CET) and we can continue our current investigation on the TellTale address space integrity. @Min Yu would you please schedule such extra session? Thanks Gab
|
|
|
|
Event: ELISA Safety-Architecture Weekly Meeting - 03/15/2022
#cal-reminder
safety-architecture@lists.elisa.tech Calendar <noreply@...>
Reminder: ELISA Safety-Architecture Weekly Meeting When: Where: Organizer: myu@... myu@... Description:
|
|
|
|
Cancelled Event: ELISA Safety-Architecture Weekly Meeting - Tuesday, March 8, 2022
#cal-cancelled
safety-architecture@lists.elisa.tech Calendar <noreply@...>
Cancelled: ELISA Safety-Architecture Weekly Meeting This event has been cancelled. When: Where: Organizer: myu@... myu@... Description:
|
|
|
|
Moving to a biweekly cadence
Hi Min Following yesterday's call we decided to move the arch wg meeting to a biweekly cadence. Could you please reschedule the sequence accordingly? The next meeting will be on Mar 15th Thanks Gab
|
|
|
|
ww9 - Agenda
Hi All For today's meeting I'll propose the following agenda. a) WG cadence: proposal to move it to a biweekly b) WG documentation: call for help to rearrange the doc c) WG short term tasks about https://docs.google.com/document/d/1JrjBHuYopU9Z7xOwm7KOc5DTKzRnvjk-NIY2QfBlBs0/edit Thanks Gab
|
|
|
|
Event: ELISA Safety-Architecture Weekly Meeting - 03/01/2022
#cal-reminder
safety-architecture@lists.elisa.tech Calendar <noreply@...>
Reminder: ELISA Safety-Architecture Weekly Meeting When: Where: Organizer: myu@... myu@... Description:
|
|
|
|
Re: ww07 - Agenda
Igor Stoppa
Hi,
> As next steps I will try to understand where and how these mappings are maintained within the Kernel. struct task_struct contains struct mm_struct *mm and struct mm_struct *active_mm which are the memory maps used by a process (and its threads) -- igor
|
|
|
|
no updates for today
Hi All Unfortunately I haven't made much progress on the telltale safety analysis compared to last week. So I'd suggest cancelling today's meeting and continuing to work offline, then we can reconvene next week. Also next week I would like to propose to move to a biweekly cadence (as it seems that we are often skipping one every two meetings...) Thanks Gab
|
|
|
|
Re: ww07 - Agenda
Hi Igor On Mon, Feb 21, 2022 at 9:00 AM Igor Stoppa via lists.elisa.tech <istoppa=nvidia.com@...> wrote: Hello, Thanks a lot for your analysis. Interestingly I have now compiled the dummy program using "gcc -nostdlib test1.c" The compilation succeeded and I ran the program; the resulting maps is: [gpaoloni@fedora test]$ cat /proc/17846/maps 00400000-00401000 r--p 00000000 00:21 9889982 /home/gpaoloni/test/a.out 00401000-00402000 r-xp 00001000 00:21 9889982 /home/gpaoloni/test/a.out 00402000-00403000 r--p 00002000 00:21 9889982 /home/gpaoloni/test/a.out 7ffde680e000-7ffde682f000 rw-p 00000000 00:00 0 [stack] 7ffde68da000-7ffde68de000 r--p 00000000 00:00 0 [vvar] 7ffde68de000-7ffde68e0000 r-xp 00000000 00:00 0 [vdso] ffffffffff600000-ffffffffff601000 --xp 00000000 00:00 0 [vsyscall] Now the mappings, as expected, do not show any glibc one. We can see a read only area where we have the code and the exception handling frame. As next steps I will try to understand where and how these mappings are maintained within the Kernel. Thanks Gab
|
|
|
|
Agenda weekly call - Feb 22 2022
Elana C
For this week's call, we are privileged to have Wenhui and Ruide from bytedance/TikTok who will help to clarify business requirements for Linux based safety critical applications. In particular,
Wenhui and Ruide will explain the technical context and invite safety experts to provide feedback on what types of considerations are relevant for these use cases.
I am adding other WGs to this meeting, so that people who have an interest in this opportunity can join and learn / provide feedback. Regards Elana
|
|
|
|
Event: ELISA Safety-Architecture Weekly Meeting - 02/22/2022
#cal-reminder
safety-architecture@lists.elisa.tech Calendar <noreply@...>
Reminder: ELISA Safety-Architecture Weekly Meeting When: Where: Organizer: myu@... myu@... Description:
|
|
|
|
Re: ww07 - Agenda
Igor Stoppa
Hello,
followup wrt identification of symbols used by an empty program, since it was pulling in libc. The program "test.c": int main() { while (1); /* Infinite loop */ return 0; } gcc test.c -o test My initial thought of using prelink did not work, because apparently prelink has been bitrotting at least since 2017. Admittedly, also yours truly had not used it since 2010. For some reason, it's still around, though. Anyways, there are various other ways to obtain the desired information. One of them is to use nm: nm -D test w __cxa_finalize@....5
w __gmon_start__
w _ITM_deregisterTMCloneTable
w _ITM_registerTMCloneTable
U __libc_start_main@...
From man nm: [...] "U" The symbol is undefined. [...] "W"
"w" The symbol is a weak symbol that has not been specifically tagged as a weak object symbol. When a weak defined symbol is linked with a normal defined symbol, the normal
defined symbol is used with no error. When a weak undefined symbol is linked and the symbol is not defined, the value of the symbol is determined in a system-specific manner
without error. On some systems, uppercase indicates that a default value has been specified. [...] Those weak housekeeping symbols seem to be the reason why libc is being pulled in. -- igor
|
|
|
|
ww07 - Agenda
Hi All Today we'll continue the investigation on the Telltale Safety App address space as per KSR_002 doc Thanks Gab
|
|
|
|
Event: ELISA Safety-Architecture Weekly Meeting - 02/15/2022
#cal-reminder
safety-architecture@lists.elisa.tech Calendar <noreply@...>
Reminder: ELISA Safety-Architecture Weekly Meeting When: Where: Organizer: myu@... myu@... Description:
|
|
|
|
Today's meeting cancelled
Hi All Apologies for the last minute notice. Today I have to cancel due to an unexpected conflict on my side. Kind Regards Gab
|
|
|
|
Event: ELISA Safety-Architecture Weekly Meeting - 02/08/2022
#cal-reminder
safety-architecture@lists.elisa.tech Calendar <noreply@...>
Reminder: ELISA Safety-Architecture Weekly Meeting When: Where: Organizer: myu@... myu@... Description:
|
|
|
|
ww5 Agenda
Hi All Today we will present the work that Daniel and myself are doing in Red Hat to qualify Kernel Critical subsystems and also claim FFI wrt the rest of the Kernel Code. The presentation will be based on the watchdog subsystem, that will serve as sample subsystem; the idea is to extend the concept to other critical subsystems (for example following the current investigation of safety app address space integrity) Thanks Gab
|
|
|
|
Event: ELISA Safety-Architecture Weekly Meeting - 02/01/2022
#cal-reminder
safety-architecture@lists.elisa.tech Calendar <noreply@...>
Reminder: ELISA Safety-Architecture Weekly Meeting When: Where: Organizer: myu@... myu@... Description:
|
|
|
|
Re: ww04 Agenda
Biasci Alessandro
Dear Gab.
in my x86 (64bit) Linux system, the implemented vdso are :
Symbol table '.dynsym' contains 12 entries: Num: Value Size Type Bind Vis Ndx Name 0: 0000000000000000 0 NOTYPE LOCAL DEFAULT UND 1: 00000000000008e0 172 FUNC WEAK DEFAULT 11 clock_gettime@@LINUX_2.6 2: 0000000000000840 136 FUNC GLOBAL DEFAULT 11 __vdso_gettimeofday@@LINUX_2.6 3: 0000000000000990 5 FUNC WEAK DEFAULT 11 clock_getres@@LINUX_2.6 4: 0000000000000990 5 FUNC GLOBAL DEFAULT 11 __vdso_clock_getres@@LINUX_2.6 5: 0000000000000840 136 FUNC WEAK DEFAULT 11 gettimeofday@@LINUX_2.6 6: 00000000000008d0 16 FUNC GLOBAL DEFAULT 11 __vdso_time@@LINUX_2.6 7: 00000000000008d0 16 FUNC WEAK DEFAULT 11 time@@LINUX_2.6 8: 00000000000008e0 172 FUNC GLOBAL DEFAULT 11 __vdso_clock_gettime@@LINUX_2.6 9: 0000000000000000 0 OBJECT GLOBAL DEFAULT ABS LINUX_2.6 10: 00000000000009a0 37 FUNC GLOBAL DEFAULT 11 __vdso_getcpu@@LINUX_2.6 11: 00000000000009a0 37 FUNC WEAK DEFAULT 11 getcpu@@LINUX_2.6
These are “per-architecture” so, on ARM, the set can be different.
B.R. Alessandro.
From: safety-architecture@... [mailto:safety-architecture@...]
On Behalf Of Gabriele Paoloni
Sent: Monday, January 24, 2022 5:39 PM To: safety-architecture@... Subject: [ELISA Safety Architecture WG] ww04 Agenda
Hi All
I am recovering and back in the trenches so I am sending out the agenda for tomorrow's meeting: 1) address space usage of the telltale safety app 2) Moving the WG docs from GDrive to github
Thanks Gab
|
|
|