Question regarding trying to avoid `perf_event_paranoid` using `setcap`

Hello there,

and apologies if the answer is obvious. I wanted to avoid using the axe

# echo 'kernel.perf_event_paranoid=1' | sudo tee '/etc/sysctl.d/51-enable-perf-events.conf'

and instead, opted for the https://unix.stackexchange.com/a/519071/266638 solution - tl;dr:

sudo newgrp perfmon_users
sudo usermod -aG perfmon_users stdedos

sudo chgrp perfmon_users /usr/bin/rr 
chmod o-rwx /usr/bin/rr
ls -lah /usr/bin/rr
sudo chmod o-rwx /usr/bin/rr
sudo setcap "cap_perfmon,cap_sys_ptrace,cap_syslog=ep" /usr/bin/rr
sudo setcap "cap_sys_ptrace,cap_syslog=ep" /usr/bin/rr
# Only this one ^^^ works on Ubuntu 20.04, ******; or
sudo setcap -v "38,cap_sys_ptrace,cap_syslog=ep" /usr/bin/rr

and I am still getting

$ rr record --wait sleep 20
[FATAL /home/roc/rr/rr/src/PerfCounters.cc:213:start_counter()] Permission denied to use 'perf_event_open'; are hardware perf events available? See https://github.com/rr-debugger/rr/wiki/Will-rr-work-on-my-system

even though, by using sudo, "it seems to be working":

$ cat /proc/1573015/status
Name:	sudo
Umask:	0022
State:	S (sleeping)
Tgid:	1573015
Ngid:	0
Pid:	1573015
PPid:	1553277
TracerPid:	0
Uid:	0	0	0	0
Gid:	0	0	0	0
FDSize:	256
Groups:	0 
NStgid:	1573015
NSpid:	1573015
NSpgid:	1573015
NSsid:	1553133
VmPeak:	   16192 kB
VmSize:	   16192 kB
VmLck:	       0 kB
VmPin:	       0 kB
VmHWM:	    4908 kB
VmRSS:	    4908 kB
RssAnon:	     704 kB
RssFile:	    4204 kB
RssShmem:	       0 kB
VmData:	     856 kB
VmStk:	     276 kB
VmExe:	     104 kB
VmLib:	    3340 kB
VmPTE:	      56 kB
VmSwap:	       0 kB
HugetlbPages:	       0 kB
CoreDumping:	0
THP_enabled:	1
Threads:	1
SigQ:	1/126851
SigPnd:	0000000000000000
ShdPnd:	0000000000000000
SigBlk:	0000000000000000
SigIgn:	0000000000000000
SigCgt:	00000001800b7a07
CapInh:	0000000000000000
CapPrm:	000001ffffffffff
CapEff:	000001ffffffffff
CapBnd:	000001ffffffffff
CapAmb:	0000000000000000
NoNewPrivs:	0
Seccomp:	0
Seccomp_filters:	0
Speculation_Store_Bypass:	thread vulnerable
SpeculationIndirectBranch:	conditional enabled
Cpus_allowed:	ffff
Cpus_allowed_list:	0-15
Mems_allowed:	00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000000,00000001
Mems_allowed_list:	0
voluntary_ctxt_switches:	2
nonvoluntary_ctxt_switches:	0
$ capsh --decode=000001ffffffffff
WARNING: libcap needs an update (cap=40 should have a name).
0x000001ffffffffff=cap_chown,cap_dac_override,cap_dac_read_search,cap_fowner,cap_fsetid,cap_kill,cap_setgid,cap_setuid,cap_setpcap,cap_linux_immutable,cap_net_bind_service,cap_net_broadcast,cap_net_admin,cap_net_raw,cap_ipc_lock,cap_ipc_owner,cap_sys_module,cap_sys_rawio,cap_sys_chroot,cap_sys_ptrace,cap_sys_pacct,cap_sys_admin,cap_sys_boot,cap_sys_nice,cap_sys_resource,cap_sys_time,cap_sys_tty_config,cap_mknod,cap_lease,cap_audit_write,cap_audit_control,cap_setfcap,cap_mac_override,cap_mac_admin,cap_syslog,cap_wake_alarm,cap_block_suspend,cap_audit_read,38,39,40

(I don't know a way to "freeze" rr while running it in a non-sudo environment; any advice welcome)

Start by reproducing rr record --wait sleep 20 with and without sudo, then inspect src/PerfCounters.cc at start_counter() and the reported perf_event_open failure. Compare the executable capabilities on /usr/bin/rr with the permissions required by the kernel; done means explaining whether this configuration is supported and identifying any rr-side change or documentation needed.