2020-08-25 18:30:26

by Nathan Royce

[permalink] [raw]
Subject: localmodconfig - "intel_rapl_perf config not found!!"

Intel Haswell
Linux 5.8.3

First time I've used localmodconfig ever since reading what it does
and liking the "supposed" kernel customization specific to the system.
I only use quotes on "supposed" because I DO still see entries I have
no interest in (not applicable to my system/needs).
I don't know if another email would be warranted for localmodconfig
only or if my expectation of it is unrealistic.

The "intel_rapl_perf config not found!!" comes up with every .config I try.
The simplest test I can come up with would be:
*****
make defconfig //x86_64_defconfig
lsmod > lsmod.cfg
make localmodconfig LSMOD=lsmod.cfg
*****

lsmod.cfg
*****
Module Size Used by
uinput 20480 1
rfcomm 94208 16
ccm 20480 9
cmac 16384 5
algif_hash 16384 2
bnep 28672 2
btrfs 1556480 1
blake2b_generic 20480 0
xor 24576 1 btrfs
raid6_pq 122880 1 btrfs
libcrc32c 16384 1 btrfs
crc32c_generic 16384 0
nls_iso8859_1 16384 1
nls_cp437 20480 1
vfat 24576 1
fat 90112 1 vfat
snd_usb_audio 311296 0
snd_usbmidi_lib 45056 1 snd_usb_audio
snd_rawmidi 45056 1 snd_usbmidi_lib
snd_seq_device 16384 1 snd_rawmidi
tda18271 53248 1
au8522_dig 16384 1
au8522_common 16384 1 au8522_dig
au0828 69632 1
tveeprom 28672 1 au0828
dvb_core 176128 1 au0828
videobuf2_vmalloc 20480 2 dvb_core,au0828
videobuf2_memops 20480 1 videobuf2_vmalloc
videobuf2_v4l2 28672 1 au0828
intel_rapl_msr 20480 0
btusb 57344 0
videodev 274432 2 videobuf2_v4l2,au0828
btrtl 24576 1 btusb
btbcm 20480 1 btusb
videobuf2_common 57344 3 videobuf2_v4l2,dvb_core,au0828
intel_rapl_common 32768 1 intel_rapl_msr
rc_core 61440 1 au0828
btintel 32768 1 btusb
bluetooth 688128 49 btrtl,btintel,btbcm,bnep,btusb,rfcomm
x86_pkg_temp_thermal 20480 0
intel_powerclamp 20480 0
mousedev 24576 0
ecdh_generic 16384 2 bluetooth
ecc 36864 1 ecdh_generic
crc16 16384 1 bluetooth
rtl8821ae 290816 0
coretemp 20480 0
snd_hda_codec_hdmi 73728 1
btcoexist 225280 1 rtl8821ae
kvm_intel 335872 0
rtl_pci 36864 1 rtl8821ae
rtlwifi 139264 3 rtl_pci,rtl8821ae,btcoexist
kvm 876544 1 kvm_intel
iTCO_wdt 16384 0
mei_hdcp 24576 0
iTCO_vendor_support 16384 1 iTCO_wdt
mac80211 954368 3 rtl_pci,rtl8821ae,rtlwifi
i915 2703360 60
snd_hda_codec_realtek 143360 1
irqbypass 16384 1 kvm
snd_hda_codec_generic 98304 1 snd_hda_codec_realtek
ledtrig_audio 16384 2 snd_hda_codec_generic,snd_hda_codec_realtek
intel_cstate 16384 0
snd_hda_intel 53248 4
snd_soc_rt5640 147456 0
intel_uncore 163840 0
cfg80211 925696 2 rtlwifi,mac80211
snd_intel_dspcfg 24576 1 snd_hda_intel
snd_soc_rl6231 20480 1 snd_soc_rt5640
intel_rapl_perf 16384 2
snd_hda_codec 176128 4
snd_hda_codec_generic,snd_hda_codec_hdmi,snd_hda_intel,snd_hda_codec_realtek
snd_soc_core 311296 1 snd_soc_rt5640
rfkill 32768 11 bluetooth,cfg80211
libarc4 16384 1 mac80211
snd_compress 32768 1 snd_soc_core
alx 57344 0
input_leds 16384 0
snd_hda_core 114688 5
snd_hda_codec_generic,snd_hda_codec_hdmi,snd_hda_intel,snd_hda_codec,snd_hda_codec_realtek
i2c_algo_bit 16384 1 i915
ac97_bus 16384 1 snd_soc_core
snd_pcm_dmaengine 16384 1 snd_soc_core
snd_hwdep 20480 2 snd_usb_audio,snd_hda_codec
mdio 16384 1 alx
drm_kms_helper 266240 1 i915
snd_pcm 159744 9
snd_hda_codec_hdmi,snd_hda_intel,snd_usb_audio,snd_hda_codec,snd_soc_rt5640,snd_compress,snd_soc_core,snd_hda_core,snd_pcm_dmaengine
cec 69632 2 drm_kms_helper,i915
snd_timer 49152 1 snd_pcm
mei_me 49152 1
intel_gtt 24576 1 i915
syscopyarea 16384 1 drm_kms_helper
sysfillrect 16384 1 drm_kms_helper
sysimgblt 16384 1 drm_kms_helper
snd 118784 22
snd_hda_codec_generic,snd_seq_device,snd_hda_codec_hdmi,snd_hwdep,snd_hda_intel,snd_usb_audio,snd_usbmidi_lib,snd_hda_codec,snd_hda_codec_realtek,snd_timer,snd_compress,snd_soc_core,snd_pcm,snd_rawmidi
fb_sys_fops 16384 1 drm_kms_helper
soundcore 16384 1 snd
mei 131072 3 mei_hdcp,mei_me
i2c_i801 36864 0
lpc_ich 28672 0
evdev 28672 12
mac_hid 16384 0
pcspkr 16384 0
drm 606208 16 drm_kms_helper,i915
agpgart 53248 2 intel_gtt,drm
ip_tables 36864 0
x_tables 53248 1 ip_tables
f2fs 630784 1
dm_crypt 49152 2
serpent_avx2 49152 4
serpent_avx_x86_64 49152 1 serpent_avx2
serpent_sse2_x86_64 53248 0
serpent_generic 32768 5
serpent_avx2,serpent_sse2_x86_64,serpent_avx_x86_64
algif_skcipher 16384 2
af_alg 36864 10 algif_hash,algif_skcipher
uas 32768 0
usb_storage 81920 3 uas
sr_mod 28672 0
sd_mod 61440 5
cdrom 77824 1 sr_mod
hid_logitech_hidpp 45056 0
hid_logitech_dj 28672 0
hid_generic 16384 0
usbhid 65536 1 hid_logitech_dj
hid 143360 4
usbhid,hid_generic,hid_logitech_dj,hid_logitech_hidpp
crct10dif_pclmul 16384 1
crc32_pclmul 16384 1
serio_raw 20480 0
crc32c_intel 24576 2
atkbd 36864 0
libps2 20480 1 atkbd
ghash_clmulni_intel 16384 0
ahci 40960 1
libahci 45056 1 ahci
libata 303104 2 libahci,ahci
aesni_intel 372736 13
scsi_mod 274432 5 sd_mod,usb_storage,uas,libata,sr_mod
crypto_simd 16384 4
serpent_avx2,serpent_sse2_x86_64,aesni_intel,serpent_avx_x86_64
cryptd 28672 6 crypto_simd,ghash_clmulni_intel
glue_helper 16384 4
serpent_avx2,serpent_sse2_x86_64,aesni_intel,serpent_avx_x86_64
xhci_pci 20480 0
xhci_hcd 294912 1 xhci_pci
ehci_pci 20480 0
ehci_hcd 98304 1 ehci_pci
i8042 32768 0
serio 28672 4 serio_raw,atkbd,i8042
dm_mod 167936 10 dm_crypt
pkcs8_key_parser 16384 0
nct6775 77824 0
hwmon_vid 16384 1 nct6775
crypto_user 16384 0
*****

That alone is problematic in that localmodconfig is only going by
what's in the .config and not filling in the blanks, so I have to
"make menuconfig" and go through each "module <*> did not have configs
<CONFIG_*>" and manually set them.

But even after I take care of all the missing entries, the
"intel_rapl_perf" still comes up when running localmodconfig again.

I'm going to assume it has something to do with the naming and it's
supposed to be associated with "Intel/AMD rapl performance events
(CONFIG_PERF_EVENTS_INTEL_RAPL)" which I already have set to 'Y'.
Right below that, I also get 'Use of uninitialized value
$ENV{"LMC_KEEP"} in split at ./scripts/kconfig/streamline_config.pl
line 596.', but again that is the sort of thing that may warrant a new
email specific to localmodconfig author(s). But again maybe not
because I take it more as a warning given I don't make use of
LMC_KEEP.


2020-08-25 19:17:04

by Randy Dunlap

[permalink] [raw]
Subject: Re: localmodconfig - "intel_rapl_perf config not found!!"

On 8/25/20 11:28 AM, Nathan Royce wrote:
> Intel Haswell
> Linux 5.8.3
>
> First time I've used localmodconfig ever since reading what it does
> and liking the "supposed" kernel customization specific to the system.
> I only use quotes on "supposed" because I DO still see entries I have
> no interest in (not applicable to my system/needs).
> I don't know if another email would be warranted for localmodconfig
> only or if my expectation of it is unrealistic.
>
> The "intel_rapl_perf config not found!!" comes up with every .config I try.
> The simplest test I can come up with would be:
> *****
> make defconfig //x86_64_defconfig
> lsmod > lsmod.cfg
> make localmodconfig LSMOD=lsmod.cfg
> *****

so intel_rapl_perf is listed in your lsmod.cfg file:
intel_rapl_perf 16384 2

You say Linux 5.8.3. I'm guessing that your "make localmodconfig" tree
is Linux 5.8.3 (?). What kernel version are you running?
I think that it's older, and some file/module names have changed since then.


>
> lsmod.cfg
> *****
> Module Size Used by
> uinput 20480 1
> rfcomm 94208 16
> ccm 20480 9
> cmac 16384 5
> algif_hash 16384 2
> bnep 28672 2
> btrfs 1556480 1
> blake2b_generic 20480 0
> xor 24576 1 btrfs
> raid6_pq 122880 1 btrfs
> libcrc32c 16384 1 btrfs
> crc32c_generic 16384 0
> nls_iso8859_1 16384 1
> nls_cp437 20480 1
> vfat 24576 1
> fat 90112 1 vfat
> snd_usb_audio 311296 0
> snd_usbmidi_lib 45056 1 snd_usb_audio
> snd_rawmidi 45056 1 snd_usbmidi_lib
> snd_seq_device 16384 1 snd_rawmidi
> tda18271 53248 1
> au8522_dig 16384 1
> au8522_common 16384 1 au8522_dig
> au0828 69632 1
> tveeprom 28672 1 au0828
> dvb_core 176128 1 au0828
> videobuf2_vmalloc 20480 2 dvb_core,au0828
> videobuf2_memops 20480 1 videobuf2_vmalloc
> videobuf2_v4l2 28672 1 au0828
> intel_rapl_msr 20480 0
> btusb 57344 0
> videodev 274432 2 videobuf2_v4l2,au0828
> btrtl 24576 1 btusb
> btbcm 20480 1 btusb
> videobuf2_common 57344 3 videobuf2_v4l2,dvb_core,au0828
> intel_rapl_common 32768 1 intel_rapl_msr
> rc_core 61440 1 au0828
> btintel 32768 1 btusb
> bluetooth 688128 49 btrtl,btintel,btbcm,bnep,btusb,rfcomm
> x86_pkg_temp_thermal 20480 0
> intel_powerclamp 20480 0
> mousedev 24576 0
> ecdh_generic 16384 2 bluetooth
> ecc 36864 1 ecdh_generic
> crc16 16384 1 bluetooth
> rtl8821ae 290816 0
> coretemp 20480 0
> snd_hda_codec_hdmi 73728 1
> btcoexist 225280 1 rtl8821ae
> kvm_intel 335872 0
> rtl_pci 36864 1 rtl8821ae
> rtlwifi 139264 3 rtl_pci,rtl8821ae,btcoexist
> kvm 876544 1 kvm_intel
> iTCO_wdt 16384 0
> mei_hdcp 24576 0
> iTCO_vendor_support 16384 1 iTCO_wdt
> mac80211 954368 3 rtl_pci,rtl8821ae,rtlwifi
> i915 2703360 60
> snd_hda_codec_realtek 143360 1
> irqbypass 16384 1 kvm
> snd_hda_codec_generic 98304 1 snd_hda_codec_realtek
> ledtrig_audio 16384 2 snd_hda_codec_generic,snd_hda_codec_realtek
> intel_cstate 16384 0
> snd_hda_intel 53248 4
> snd_soc_rt5640 147456 0
> intel_uncore 163840 0
> cfg80211 925696 2 rtlwifi,mac80211
> snd_intel_dspcfg 24576 1 snd_hda_intel
> snd_soc_rl6231 20480 1 snd_soc_rt5640
> intel_rapl_perf 16384 2
> snd_hda_codec 176128 4
> snd_hda_codec_generic,snd_hda_codec_hdmi,snd_hda_intel,snd_hda_codec_realtek
> snd_soc_core 311296 1 snd_soc_rt5640
> rfkill 32768 11 bluetooth,cfg80211
> libarc4 16384 1 mac80211
> snd_compress 32768 1 snd_soc_core
> alx 57344 0
> input_leds 16384 0
> snd_hda_core 114688 5
> snd_hda_codec_generic,snd_hda_codec_hdmi,snd_hda_intel,snd_hda_codec,snd_hda_codec_realtek
> i2c_algo_bit 16384 1 i915
> ac97_bus 16384 1 snd_soc_core
> snd_pcm_dmaengine 16384 1 snd_soc_core
> snd_hwdep 20480 2 snd_usb_audio,snd_hda_codec
> mdio 16384 1 alx
> drm_kms_helper 266240 1 i915
> snd_pcm 159744 9
> snd_hda_codec_hdmi,snd_hda_intel,snd_usb_audio,snd_hda_codec,snd_soc_rt5640,snd_compress,snd_soc_core,snd_hda_core,snd_pcm_dmaengine
> cec 69632 2 drm_kms_helper,i915
> snd_timer 49152 1 snd_pcm
> mei_me 49152 1
> intel_gtt 24576 1 i915
> syscopyarea 16384 1 drm_kms_helper
> sysfillrect 16384 1 drm_kms_helper
> sysimgblt 16384 1 drm_kms_helper
> snd 118784 22
> snd_hda_codec_generic,snd_seq_device,snd_hda_codec_hdmi,snd_hwdep,snd_hda_intel,snd_usb_audio,snd_usbmidi_lib,snd_hda_codec,snd_hda_codec_realtek,snd_timer,snd_compress,snd_soc_core,snd_pcm,snd_rawmidi
> fb_sys_fops 16384 1 drm_kms_helper
> soundcore 16384 1 snd
> mei 131072 3 mei_hdcp,mei_me
> i2c_i801 36864 0
> lpc_ich 28672 0
> evdev 28672 12
> mac_hid 16384 0
> pcspkr 16384 0
> drm 606208 16 drm_kms_helper,i915
> agpgart 53248 2 intel_gtt,drm
> ip_tables 36864 0
> x_tables 53248 1 ip_tables
> f2fs 630784 1
> dm_crypt 49152 2
> serpent_avx2 49152 4
> serpent_avx_x86_64 49152 1 serpent_avx2
> serpent_sse2_x86_64 53248 0
> serpent_generic 32768 5
> serpent_avx2,serpent_sse2_x86_64,serpent_avx_x86_64
> algif_skcipher 16384 2
> af_alg 36864 10 algif_hash,algif_skcipher
> uas 32768 0
> usb_storage 81920 3 uas
> sr_mod 28672 0
> sd_mod 61440 5
> cdrom 77824 1 sr_mod
> hid_logitech_hidpp 45056 0
> hid_logitech_dj 28672 0
> hid_generic 16384 0
> usbhid 65536 1 hid_logitech_dj
> hid 143360 4
> usbhid,hid_generic,hid_logitech_dj,hid_logitech_hidpp
> crct10dif_pclmul 16384 1
> crc32_pclmul 16384 1
> serio_raw 20480 0
> crc32c_intel 24576 2
> atkbd 36864 0
> libps2 20480 1 atkbd
> ghash_clmulni_intel 16384 0
> ahci 40960 1
> libahci 45056 1 ahci
> libata 303104 2 libahci,ahci
> aesni_intel 372736 13
> scsi_mod 274432 5 sd_mod,usb_storage,uas,libata,sr_mod
> crypto_simd 16384 4
> serpent_avx2,serpent_sse2_x86_64,aesni_intel,serpent_avx_x86_64
> cryptd 28672 6 crypto_simd,ghash_clmulni_intel
> glue_helper 16384 4
> serpent_avx2,serpent_sse2_x86_64,aesni_intel,serpent_avx_x86_64
> xhci_pci 20480 0
> xhci_hcd 294912 1 xhci_pci
> ehci_pci 20480 0
> ehci_hcd 98304 1 ehci_pci
> i8042 32768 0
> serio 28672 4 serio_raw,atkbd,i8042
> dm_mod 167936 10 dm_crypt
> pkcs8_key_parser 16384 0
> nct6775 77824 0
> hwmon_vid 16384 1 nct6775
> crypto_user 16384 0
> *****
>
> That alone is problematic in that localmodconfig is only going by
> what's in the .config and not filling in the blanks, so I have to
> "make menuconfig" and go through each "module <*> did not have configs
> <CONFIG_*>" and manually set them.
>
> But even after I take care of all the missing entries, the
> "intel_rapl_perf" still comes up when running localmodconfig again.
>
> I'm going to assume it has something to do with the naming and it's
> supposed to be associated with "Intel/AMD rapl performance events
> (CONFIG_PERF_EVENTS_INTEL_RAPL)" which I already have set to 'Y'.
> Right below that, I also get 'Use of uninitialized value
> $ENV{"LMC_KEEP"} in split at ./scripts/kconfig/streamline_config.pl
> line 596.', but again that is the sort of thing that may warrant a new
> email specific to localmodconfig author(s). But again maybe not
> because I take it more as a warning given I don't make use of
> LMC_KEEP.
>


--
~Randy

2020-08-25 20:36:42

by Nathan Royce

[permalink] [raw]
Subject: Re: localmodconfig - "intel_rapl_perf config not found!!"

Correct. I'm building for 5.8.3 and I'm currently on 5.7.4 (1 month
doesn't seem particularly old).

On Tue, Aug 25, 2020 at 2:13 PM Randy Dunlap <[email protected]> wrote:
>
> so intel_rapl_perf is listed in your lsmod.cfg file:
> intel_rapl_perf 16384 2
>
> You say Linux 5.8.3. I'm guessing that your "make localmodconfig" tree
> is Linux 5.8.3 (?). What kernel version are you running?
> I think that it's older, and some file/module names have changed since then.

2020-08-25 21:13:54

by Randy Dunlap

[permalink] [raw]
Subject: Re: localmodconfig - "intel_rapl_perf config not found!!"

> On Tue, Aug 25, 2020 at 2:13 PM Randy Dunlap <[email protected]> wrote:
>>
>> so intel_rapl_perf is listed in your lsmod.cfg file:
>> intel_rapl_perf 16384 2
>>
>> You say Linux 5.8.3. I'm guessing that your "make localmodconfig" tree
>> is Linux 5.8.3 (?). What kernel version are you running?
>> I think that it's older, and some file/module names have changed since then.

On 8/25/20 1:34 PM, Nathan Royce wrote:
> Correct. I'm building for 5.8.3 and I'm currently on 5.7.4 (1 month
> doesn't seem particularly old).

Yes, things can change quickly.


I don't see any support in streamline_config.pl for Kconfig symbols
and/or modules whose names have changed. Trying to do something
like that would be a never-ending job (a la job security).

At least it gave you a warning that it couldn't find a Kconfig symbol
for that module.


From your original email:
| I'm going to assume it has something to do with the naming and it's
| supposed to be associated with "Intel/AMD rapl performance events
| (CONFIG_PERF_EVENTS_INTEL_RAPL)" which I already have set to 'Y'.

Yes, commit fd3ae1e1587d64ef8cc8e361903d33625458073e changed the module name
since it now supports both Intel and AMD.


| Right below that, I also get 'Use of uninitialized value
| $ENV{"LMC_KEEP"} in split at ./scripts/kconfig/streamline_config.pl
| line 596.', but again that is the sort of thing that may warrant a new
| email specific to localmodconfig author(s). But again maybe not
| because I take it more as a warning given I don't make use of
| LMC_KEEP.


@Changbin: can you fix this little bug in streamline_config.pl, please? ^^^^^


--
~Randy

2020-08-25 21:35:53

by Randy Dunlap

[permalink] [raw]
Subject: Re: localmodconfig - "intel_rapl_perf config not found!!"

On 8/25/20 2:12 PM, Randy Dunlap wrote:
>> On Tue, Aug 25, 2020 at 2:13 PM Randy Dunlap <[email protected]> wrote:
>>>
>>> so intel_rapl_perf is listed in your lsmod.cfg file:
>>> intel_rapl_perf 16384 2
>>>
>>> You say Linux 5.8.3. I'm guessing that your "make localmodconfig" tree
>>> is Linux 5.8.3 (?). What kernel version are you running?
>>> I think that it's older, and some file/module names have changed since then.
>
> On 8/25/20 1:34 PM, Nathan Royce wrote:
>> Correct. I'm building for 5.8.3 and I'm currently on 5.7.4 (1 month
>> doesn't seem particularly old).
>
> Yes, things can change quickly.
>
>
> I don't see any support in streamline_config.pl for Kconfig symbols
> and/or modules whose names have changed. Trying to do something
> like that would be a never-ending job (a la job security).
>
> At least it gave you a warning that it couldn't find a Kconfig symbol
> for that module.
>
>
> From your original email:
> | I'm going to assume it has something to do with the naming and it's
> | supposed to be associated with "Intel/AMD rapl performance events
> | (CONFIG_PERF_EVENTS_INTEL_RAPL)" which I already have set to 'Y'.
>
> Yes, commit fd3ae1e1587d64ef8cc8e361903d33625458073e changed the module name
> since it now supports both Intel and AMD.
>
>
> | Right below that, I also get 'Use of uninitialized value
> | $ENV{"LMC_KEEP"} in split at ./scripts/kconfig/streamline_config.pl
> | line 596.', but again that is the sort of thing that may warrant a new
> | email specific to localmodconfig author(s). But again maybe not
> | because I take it more as a warning given I don't make use of
> | LMC_KEEP.
>
>
> @Changbin: can you fix this little bug in streamline_config.pl, please? ^^^^^

Changing email address for Changbin.
Plus I made the quick patch for streamline_config.pl (below)
although I haven't tested it.

---
From: Randy Dunlap <[email protected]>

A user reported:
'Use of uninitialized value $ENV{"LMC_KEEP"} in split at
./scripts/kconfig/streamline_config.pl line 596.'

so first check that $ENV{LMC_KEEP} is defined before trying
to use it.

Fixes: c027b02d89fd ("streamline_config.pl: add LMC_KEEP to preserve some kconfigs")
Signed-off-by: Randy Dunlap <[email protected]>
Cc: Changbin Du <[email protected]>
Cc: Steven Rostedt (VMware) <[email protected]>
Cc: Masahiro Yamada <[email protected]>
---
scripts/kconfig/streamline_config.pl | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)

--- linux-next-20200825.orig/scripts/kconfig/streamline_config.pl
+++ linux-next-20200825/scripts/kconfig/streamline_config.pl
@@ -593,7 +593,10 @@ while ($repeat) {
}

my %setconfigs;
-my @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
+my @preserved_kconfigs;
+if (defined($ENV{'LMC_KEEP'})) {
+ @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
+}

sub in_preserved_kconfigs {
my $kconfig = $config2kfile{$_[0]};

2020-08-26 16:32:55

by Changbin Du

[permalink] [raw]
Subject: Re: localmodconfig - "intel_rapl_perf config not found!!"

On Tue, Aug 25, 2020 at 02:34:34PM -0700, Randy Dunlap wrote:
> On 8/25/20 2:12 PM, Randy Dunlap wrote:
> >> On Tue, Aug 25, 2020 at 2:13 PM Randy Dunlap <[email protected]> wrote:
> >>>
> >>> so intel_rapl_perf is listed in your lsmod.cfg file:
> >>> intel_rapl_perf 16384 2
> >>>
> >>> You say Linux 5.8.3. I'm guessing that your "make localmodconfig" tree
> >>> is Linux 5.8.3 (?). What kernel version are you running?
> >>> I think that it's older, and some file/module names have changed since then.
> >
> > On 8/25/20 1:34 PM, Nathan Royce wrote:
> >> Correct. I'm building for 5.8.3 and I'm currently on 5.7.4 (1 month
> >> doesn't seem particularly old).
> >
> > Yes, things can change quickly.
> >
> >
> > I don't see any support in streamline_config.pl for Kconfig symbols
> > and/or modules whose names have changed. Trying to do something
> > like that would be a never-ending job (a la job security).
> >
> > At least it gave you a warning that it couldn't find a Kconfig symbol
> > for that module.
> >
> >
> > From your original email:
> > | I'm going to assume it has something to do with the naming and it's
> > | supposed to be associated with "Intel/AMD rapl performance events
> > | (CONFIG_PERF_EVENTS_INTEL_RAPL)" which I already have set to 'Y'.
> >
> > Yes, commit fd3ae1e1587d64ef8cc8e361903d33625458073e changed the module name
> > since it now supports both Intel and AMD.
> >
> >
> > | Right below that, I also get 'Use of uninitialized value
> > | $ENV{"LMC_KEEP"} in split at ./scripts/kconfig/streamline_config.pl
> > | line 596.', but again that is the sort of thing that may warrant a new
> > | email specific to localmodconfig author(s). But again maybe not
> > | because I take it more as a warning given I don't make use of
> > | LMC_KEEP.
> >
> >
> > @Changbin: can you fix this little bug in streamline_config.pl, please? ^^^^^
This was tested before. Is this a perl version issue?

[email protected]:~$ cat a.pl
my @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
print("@preserved_kconfigs")
[email protected]:~$ perl a.pl
[email protected]:~$ perl -v
This is perl 5, version 26, subversion 1 (v5.26.1) built for x86_64-linux-gnu-thread-multi
(with 67 registered patches, see perl -V for more detail)

Copyright 1987-2017, Larry Wall

Perl may be copied only under the terms of either the Artistic License or the
GNU General Public License, which may be found in the Perl 5 source kit.

Complete documentation for Perl, including FAQ lists, should be found on
this system using "man perl" or "perldoc perl". If you have access to the
Internet, point your browser at http://www.perl.org/, the Perl Home Page.
>
> Changing email address for Changbin.
> Plus I made the quick patch for streamline_config.pl (below)
> although I haven't tested it.
>
> ---
> From: Randy Dunlap <[email protected]>
>
> A user reported:
> 'Use of uninitialized value $ENV{"LMC_KEEP"} in split at
> ./scripts/kconfig/streamline_config.pl line 596.'
>
> so first check that $ENV{LMC_KEEP} is defined before trying
> to use it.
>
> Fixes: c027b02d89fd ("streamline_config.pl: add LMC_KEEP to preserve some kconfigs")
> Signed-off-by: Randy Dunlap <[email protected]>
> Cc: Changbin Du <[email protected]>
> Cc: Steven Rostedt (VMware) <[email protected]>
> Cc: Masahiro Yamada <[email protected]>
> ---
> scripts/kconfig/streamline_config.pl | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> --- linux-next-20200825.orig/scripts/kconfig/streamline_config.pl
> +++ linux-next-20200825/scripts/kconfig/streamline_config.pl
> @@ -593,7 +593,10 @@ while ($repeat) {
> }
>
> my %setconfigs;
> -my @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
> +my @preserved_kconfigs;
> +if (defined($ENV{'LMC_KEEP'})) {
> + @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
> +}
>
> sub in_preserved_kconfigs {
> my $kconfig = $config2kfile{$_[0]};
>

--
Cheers,
Changbin Du

2020-08-26 21:11:11

by Steven Rostedt

[permalink] [raw]
Subject: Re: localmodconfig - "intel_rapl_perf config not found!!"

On Tue, 25 Aug 2020 14:34:34 -0700
Randy Dunlap <[email protected]> wrote:

> On 8/25/20 2:12 PM, Randy Dunlap wrote:
> >> On Tue, Aug 25, 2020 at 2:13 PM Randy Dunlap <[email protected]> wrote:
> >>>
> >>> so intel_rapl_perf is listed in your lsmod.cfg file:
> >>> intel_rapl_perf 16384 2
> >>>
> >>> You say Linux 5.8.3. I'm guessing that your "make localmodconfig" tree
> >>> is Linux 5.8.3 (?). What kernel version are you running?
> >>> I think that it's older, and some file/module names have changed since then.
> >
> > On 8/25/20 1:34 PM, Nathan Royce wrote:
> >> Correct. I'm building for 5.8.3 and I'm currently on 5.7.4 (1 month
> >> doesn't seem particularly old).
> >
> > Yes, things can change quickly.

Yes and as Randy said, there's no way to easily fix that.

> >
> >
> > I don't see any support in streamline_config.pl for Kconfig symbols
> > and/or modules whose names have changed. Trying to do something
> > like that would be a never-ending job (a la job security).
> >
> > At least it gave you a warning that it couldn't find a Kconfig symbol
> > for that module.

Correct, this is exactly why I have it warn. Because this happens all
the time. In most cases, I simply ignore the warning ;-)


> >
> >
> > From your original email:
> > | I'm going to assume it has something to do with the naming and it's
> > | supposed to be associated with "Intel/AMD rapl performance events
> > | (CONFIG_PERF_EVENTS_INTEL_RAPL)" which I already have set to 'Y'.
> >
> > Yes, commit fd3ae1e1587d64ef8cc8e361903d33625458073e changed the module name
> > since it now supports both Intel and AMD.
> >
> >
> > | Right below that, I also get 'Use of uninitialized value
> > | $ENV{"LMC_KEEP"} in split at ./scripts/kconfig/streamline_config.pl
> > | line 596.', but again that is the sort of thing that may warrant a new
> > | email specific to localmodconfig author(s). But again maybe not
> > | because I take it more as a warning given I don't make use of
> > | LMC_KEEP.
> >
> >
> > @Changbin: can you fix this little bug in streamline_config.pl, please? ^^^^^
>
> Changing email address for Changbin.
> Plus I made the quick patch for streamline_config.pl (below)
> although I haven't tested it.
>
> ---
> From: Randy Dunlap <[email protected]>
>
> A user reported:
> 'Use of uninitialized value $ENV{"LMC_KEEP"} in split at
> ./scripts/kconfig/streamline_config.pl line 596.'
>
> so first check that $ENV{LMC_KEEP} is defined before trying
> to use it.

Acked-by: Steven Rostedt (VMware) <[email protected]>

Randy, please send this as a stand alone patch to the KBUILD
maintainers.

-- Steve


>
> Fixes: c027b02d89fd ("streamline_config.pl: add LMC_KEEP to preserve some kconfigs")
> Signed-off-by: Randy Dunlap <[email protected]>
> Cc: Changbin Du <[email protected]>
> Cc: Steven Rostedt (VMware) <[email protected]>
> Cc: Masahiro Yamada <[email protected]>
> ---
> scripts/kconfig/streamline_config.pl | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> --- linux-next-20200825.orig/scripts/kconfig/streamline_config.pl
> +++ linux-next-20200825/scripts/kconfig/streamline_config.pl
> @@ -593,7 +593,10 @@ while ($repeat) {
> }
>
> my %setconfigs;
> -my @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
> +my @preserved_kconfigs;
> +if (defined($ENV{'LMC_KEEP'})) {
> + @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
> +}
>
> sub in_preserved_kconfigs {
> my $kconfig = $config2kfile{$_[0]};

2020-08-26 21:14:11

by Steven Rostedt

[permalink] [raw]
Subject: Re: localmodconfig - "intel_rapl_perf config not found!!"

On Wed, 26 Aug 2020 16:30:28 +0000
Changbin Du <[email protected]> wrote:

> > >
> > > @Changbin: can you fix this little bug in streamline_config.pl, please? ^^^^^
> This was tested before. Is this a perl version issue?
>
> [email protected]:~$ cat a.pl
> my @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
> print("@preserved_kconfigs")
> [email protected]:~$ perl a.pl
> [email protected]:~$ perl -v
> This is perl 5, version 26, subversion 1 (v5.26.1) built for x86_64-linux-gnu-thread-multi
> (with 67 registered patches, see perl -V for more detail)
>
> Copyright 1987-2017, Larry Wall
>
> Perl may be copied only under the terms of either the Artistic License or the
> GNU General Public License, which may be found in the Perl 5 source kit.
>
> Complete documentation for Perl, including FAQ lists, should be found on
> this system using "man perl" or "perldoc perl". If you have access to the
> Internet, point your browser at http://www.perl.org/, the Perl Home Page.

I see that warning too:

$ perl -v

This is perl 5, version 30, subversion 2 (v5.30.2) built for x86_64-linux-thread-multi
(with 70 registered patches, see perl -V for more detail)

-- Steve

2020-09-02 16:46:03

by Masahiro Yamada

[permalink] [raw]
Subject: Re: localmodconfig - "intel_rapl_perf config not found!!"

On Thu, Aug 27, 2020 at 1:30 AM Changbin Du <[email protected]> wrote:
>
> On Tue, Aug 25, 2020 at 02:34:34PM -0700, Randy Dunlap wrote:
> > On 8/25/20 2:12 PM, Randy Dunlap wrote:
> > >> On Tue, Aug 25, 2020 at 2:13 PM Randy Dunlap <[email protected]> wrote:
> > >>>
> > >>> so intel_rapl_perf is listed in your lsmod.cfg file:
> > >>> intel_rapl_perf 16384 2
> > >>>
> > >>> You say Linux 5.8.3. I'm guessing that your "make localmodconfig" tree
> > >>> is Linux 5.8.3 (?). What kernel version are you running?
> > >>> I think that it's older, and some file/module names have changed since then.
> > >
> > > On 8/25/20 1:34 PM, Nathan Royce wrote:
> > >> Correct. I'm building for 5.8.3 and I'm currently on 5.7.4 (1 month
> > >> doesn't seem particularly old).
> > >
> > > Yes, things can change quickly.
> > >
> > >
> > > I don't see any support in streamline_config.pl for Kconfig symbols
> > > and/or modules whose names have changed. Trying to do something
> > > like that would be a never-ending job (a la job security).
> > >
> > > At least it gave you a warning that it couldn't find a Kconfig symbol
> > > for that module.
> > >
> > >
> > > From your original email:
> > > | I'm going to assume it has something to do with the naming and it's
> > > | supposed to be associated with "Intel/AMD rapl performance events
> > > | (CONFIG_PERF_EVENTS_INTEL_RAPL)" which I already have set to 'Y'.
> > >
> > > Yes, commit fd3ae1e1587d64ef8cc8e361903d33625458073e changed the module name
> > > since it now supports both Intel and AMD.
> > >
> > >
> > > | Right below that, I also get 'Use of uninitialized value
> > > | $ENV{"LMC_KEEP"} in split at ./scripts/kconfig/streamline_config.pl
> > > | line 596.', but again that is the sort of thing that may warrant a new
> > > | email specific to localmodconfig author(s). But again maybe not
> > > | because I take it more as a warning given I don't make use of
> > > | LMC_KEEP.
> > >
> > >
> > > @Changbin: can you fix this little bug in streamline_config.pl, please? ^^^^^
> This was tested before. Is this a perl version issue?
>
> [email protected]:~$ cat a.pl
> my @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
> print("@preserved_kconfigs")
> [email protected]:~$ perl a.pl
> [email protected]:~$ perl -v
> This is perl 5, version 26, subversion 1 (v5.26.1) built for x86_64-linux-gnu-thread-multi
> (with 67 registered patches, see perl -V for more detail)
>
> Copyright 1987-2017, Larry Wall
>
> Perl may be copied only under the terms of either the Artistic License or the
> GNU General Public License, which may be found in the Perl 5 source kit.
>
> Complete documentation for Perl, including FAQ lists, should be found on
> this system using "man perl" or "perldoc perl". If you have access to the
> Internet, point your browser at http://www.perl.org/, the Perl Home Page.
> >
> > Changing email address for Changbin.
> > Plus I made the quick patch for streamline_config.pl (below)
> > although I haven't tested it.
> >
> > ---
> > From: Randy Dunlap <[email protected]>
> >
> > A user reported:
> > 'Use of uninitialized value $ENV{"LMC_KEEP"} in split at
> > ./scripts/kconfig/streamline_config.pl line 596.'
> >
> > so first check that $ENV{LMC_KEEP} is defined before trying
> > to use it.
> >
> > Fixes: c027b02d89fd ("streamline_config.pl: add LMC_KEEP to preserve some kconfigs")
> > Signed-off-by: Randy Dunlap <[email protected]>
> > Cc: Changbin Du <[email protected]>
> > Cc: Steven Rostedt (VMware) <[email protected]>
> > Cc: Masahiro Yamada <[email protected]>
> > ---
> > scripts/kconfig/streamline_config.pl | 5 ++++-
> > 1 file changed, 4 insertions(+), 1 deletion(-)
> >
> > --- linux-next-20200825.orig/scripts/kconfig/streamline_config.pl
> > +++ linux-next-20200825/scripts/kconfig/streamline_config.pl
> > @@ -593,7 +593,10 @@ while ($repeat) {
> > }
> >
> > my %setconfigs;
> > -my @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
> > +my @preserved_kconfigs;
> > +if (defined($ENV{'LMC_KEEP'})) {
> > + @preserved_kconfigs = split(/:/,$ENV{LMC_KEEP});
> > +}
> >
> > sub in_preserved_kconfigs {
> > my $kconfig = $config2kfile{$_[0]};
> >
>
> --
> Cheers,
> Changbin Du



Hmm, I think I did not see this warning before,
but I am not sure...
Now I see this warning too, maybe perl version-dependent.


--
Best Regards
Masahiro Yamada