[Bug jit/100207] New: Error in build()

[Bug jit/100207] New: Error in build()

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

            Bug ID: 100207
           Summary: Error in build()
           Product: gcc
           Version: 10.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: jit
          Assignee: dmalcolm at gcc dot gnu.org
          Reporter: sujay1844 at protonmail dot com
  Target Milestone: ---

Created attachment 50654
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=50654&action=edit
The output when I try to instal libgccjit

I'm getting the following error when I try to install the AUR package using
paru -S libgccjit. I thought the issue was with paru but then I tried using yay
and got the same error. Here are the last few lines of the log.



../../../gcc-10.2.0/libcpp/expr.c: In function ‘unsigned int
cpp_classify_number(cpp_reader*, const cpp_token*, const char**, location_t)’:
../../../gcc-10.2.0/libcpp/expr.c:802:18: error: format not a string literal
and no format arguments [-Werror=format-security]
  802 |        0, message);
      |                  ^
../../../gcc-10.2.0/libcpp/expr.c:805:39: error: format not a string literal
and no format arguments [-Werror=format-security]
  805 |           virtual_location, 0, message);
      |                                       ^
cc1plus: some warnings being treated as errors
make[2]: *** [Makefile:224: expr.o] Error 1
make[2]: Leaving directory
'/home/sujay1844/.cache/paru/clone/libgccjit/src/libgccjit-build/build-x86_64-pc-linux-gnu/libcpp'
make[1]: *** [Makefile:2850: all-build-libcpp] Error 2
make[1]: Leaving directory
'/home/sujay1844/.cache/paru/clone/libgccjit/src/libgccjit-build'
make: *** [Makefile:940: all] Error 2
==> ERROR: A failure occurred in build().
    Aborting...
:: Packages failed to build: libgccjit-10.2.0-2
sujay1844@Stormbreaker ~ [1]>



This is my system config



System:    Kernel: 5.11.15-arch1-2 x86_64 bits: 64 compiler: gcc v: 10.2.0 
           Desktop: sway 1.6 Distro: Arch Linux 
Machine:   Type: Laptop System: Acer product: Swift SF314-42 v: V1.09 serial:
<filter> 
           Mobo: RO model: Kona_RN v: V1.09 serial: <filter> UEFI: Insyde v:
1.09 
           date: 11/18/2020 
Battery:   ID-1: BAT1 charge: 18.3 Wh (47.3%) condition: 38.7/47.9 Wh (80.8%)
volts: 11.3 
           min: 11.4 model: Murata AP18C4K status: Discharging 
CPU:       Info: 6-Core model: AMD Ryzen 5 4500U with Radeon Graphics bits: 64
type: MCP 
           arch: Zen 2 rev: 1 cache: L2: 3 MiB 
           flags: avx avx2 lm nx pae sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3
svm 
           bogomips: 28462 
           Speed: 1397 MHz min/max: 1400/2375 MHz boost: enabled Core speeds
(MHz): 1: 1397 
           2: 1397 3: 1397 4: 1424 5: 1396 6: 1397 
Graphics:  Device-1: AMD Renoir vendor: Acer Incorporated ALI driver: amdgpu v:
kernel 
           bus-ID: 03:00.0 
           Device-2: Quanta HD User Facing type: USB driver: uvcvideo bus-ID:
1-3:2 
           Display: wayland server: X.Org 1.21.1.1 compositor: sway driver:
loaded: amdgpu 
           unloaded: modesetting,vesa resolution: 1920x1080~60Hz 
           Message: Unable to show advanced data. Required tool glxinfo
missing. 
Audio:     Device-1: AMD vendor: Acer Incorporated ALI driver: snd_hda_intel v:
kernel 
           bus-ID: 03:00.1 
           Device-2: AMD Raven/Raven2/FireFlight/Renoir Audio Processor 
           vendor: Acer Incorporated ALI driver: snd_rn_pci_acp3x v: kernel
bus-ID: 03:00.5 
           Device-3: AMD Family 17h HD Audio vendor: Acer Incorporated ALI 
           driver: snd_hda_intel v: kernel bus-ID: 03:00.6 
           Sound Server-1: ALSA v: k5.11.15-arch1-2 running: yes 
           Sound Server-2: sndio v: N/A running: no 
           Sound Server-3: JACK v: 0.125.0 running: no 
           Sound Server-4: PulseAudio v: 14.2 running: yes 
           Sound Server-5: PipeWire v: 0.3.25 running: yes 
Network:   Device-1: Intel Wi-Fi 6 AX200 driver: iwlwifi v: kernel bus-ID:
01:00.0 
           IF: wlp1s0 state: up mac: <filter> 
Bluetooth: Device-1: Intel AX200 Bluetooth type: USB driver: btusb v: 0.8
bus-ID: 1-4:3 
           Report: rfkill ID: hci0 rfk-id: 3 state: up address: see
--recommends 
Drives:    Local Storage: total: 476.94 GiB used: 86.2 GiB (18.1%) 
           ID-1: /dev/nvme0n1 vendor: SK Hynix model: HFM512GDJTNI-82A0A size:
476.94 GiB 
           temp: 34.9 C 
Partition: ID-1: / size: 216.88 GiB used: 86.06 GiB (39.7%) fs: ext4 dev:
/dev/nvme0n1p6 
           ID-2: /boot size: 511 MiB used: 149.5 MiB (29.3%) fs: vfat dev:
/dev/nvme0n1p4 
Swap:      ID-1: swap-1 type: partition size: 15.87 GiB used: 768 KiB (0.0%) 
           dev: /dev/nvme0n1p5 
Sensors:   System Temperatures: cpu: 42.2 C mobo: 0 C gpu: amdgpu temp: 41.0 C 
           Fan Speeds (RPM): N/A 
Info:      Processes: 239 Uptime: 44m Memory: 7.21 GiB used: 2.51 GiB (34.8%)
Init: systemd 
           Compilers: gcc: 10.2.0 clang: 11.1.0 Packages: 1372 Shell: fish v:
3.2.1 
           inxi: 3.3.04

[Bug jit/100207] Error in build()

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |marxin at gcc dot gnu.org

--- Comment #1 from Martin Liška <marxin at gcc dot gnu.org> ---
You're using a non-default -Wformat-security which is something GCC cannot be
built with.

[Bug jit/100207] Error in build()

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #2 from sujay1844 at protonmail dot com ---
So is the AUR package having a bug??

[Bug jit/100207] Error in build()

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #3 from Martin Liška <marxin at gcc dot gnu.org> ---
(In reply to sujay1844 from comment #2)
> So is the AUR package having a bug??

What's AUR? Can you investigate what sets the -Wformat-security ?

[Bug jit/100207] Error in build()

[dmalcolm at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #4 from David Malcolm <dmalcolm at gcc dot gnu.org> ---
(In reply to Martin Liška from comment #3)
> (In reply to sujay1844 from comment #2)
> > So is the AUR package having a bug??
> 
> What's AUR? Can you investigate what sets the -Wformat-security ?

Presumably it's this:
  https://wiki.archlinux.org/index.php/Arch_User_Repository
Sounds like AUR is injecting -Werror=format-security.

The strings in question come from:

          const char *message = CPP_OPTION (pfile, cplusplus) 
                                ? N_("use of C++11 long long integer constant")
                                : N_("use of C99 long long integer constant");

and:

          const char *message = (result & CPP_N_UNSIGNED) == CPP_N_UNSIGNED
                                ? N_("use of C++23 %<size_t%> integer
constant")
                                : N_("use of C++23 %<make_signed_t<size_t>%>
integer constant");

which appear to be safe to me.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[dmalcolm at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

David Malcolm <dmalcolm at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|Error in build()            |-Werror=format-security
                   |                            |error in AUR build of
                   |                            |libgccjit

--- Comment #5 from David Malcolm <dmalcolm at gcc dot gnu.org> ---
Presumably the AUR package of libgccjit is here:
  https://aur.archlinux.org/packages/libgccjit/

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #6 from sujay1844 at protonmail dot com ---
(In reply to David Malcolm from comment #5)
> Presumably the AUR package of libgccjit is here:
>   https://aur.archlinux.org/packages/libgccjit/

So how do I prevent the AUR package from setting -Wformat-security?? Do i need
to edit the PKGBUILD or some C++ files?? I'm not familliar with C++.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #7 from sujay1844 at protonmail dot com ---
(In reply to David Malcolm from comment #5)
> Presumably the AUR package of libgccjit is here:
>   https://aur.archlinux.org/packages/libgccjit/

So how do I prevent the AUR package from setting -Wformat-security?? Do i need
to edit the PKGBUILD or some C++ files?? I'm not familliar with C++.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #8 from sujay1844 at protonmail dot com ---
(In reply to David Malcolm from comment #5)
> Presumably the AUR package of libgccjit is here:
>   https://aur.archlinux.org/packages/libgccjit/

So how do I prevent the AUR package from setting -Wformat-security?? Do i need
to edit the PKGBUILD or some C++ files?? I'm not familliar with C++.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #9 from sujay1844 at protonmail dot com ---
Oops sorry for sending it thrice. Bugzilla redirected me to some other bug
report, so I clicked back. I thought my comment wasn't submitted, so I sent it
again. 
TLDR: i forgot to refresh my webpage and ending up thinking my comment wasn't
sent.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[dmalcolm at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

David Malcolm <dmalcolm at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |WORKSFORME
             Status|UNCONFIRMED                 |RESOLVED

--- Comment #10 from David Malcolm <dmalcolm at gcc dot gnu.org> ---
(In reply to sujay1844 from comment #8)
> (In reply to David Malcolm from comment #5)
> > Presumably the AUR package of libgccjit is here:
> >   https://aur.archlinux.org/packages/libgccjit/
> 
> So how do I prevent the AUR package from setting -Wformat-security?? Do i
> need to edit the PKGBUILD or some C++ files?? I'm not familliar with C++.

Sorry; I have no idea.  Perhaps it's inheriting a default CFLAGS or CXXFLAGS
from some AUR/PKGBUILD policy file?

This seems to be purely an issue with downstream packaging of libgccjit, so I'm
going to mark this as "RESOLVED WORKSFORME"; I think you need to file this in a
downstream AUR bug tracker if it's still not working for you.

Hope this is helpful; good luck.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

sujay1844 at protonmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|WORKSFORME                  |FIXED

--- Comment #11 from sujay1844 at protonmail dot com ---
Ok hey after some digging I found this in my /etc/makepkg.conf What should I
change?? (I'm not familiar with C++) 


#-- Compiler and Linker Flags
#CPPFLAGS=""
CFLAGS="-march=x86-64 -mtune=generic -O2 -pipe -fno-plt -fexceptions \
        -Wp,-D_FORTIFY_SOURCE=2,-D_GLIBCXX_ASSERTIONS \
        -Wformat -Werror=format-security \
        -fstack-clash-protection -fcf-protection"
CXXFLAGS="$CFLAGS"
LDFLAGS="-Wl,-O1,--sort-common,--as-needed,-z,relro,-z,now"
#RUSTFLAGS="-C opt-level=2"
#-- Make Flags: change this for DistCC/SMP systems
#MAKEFLAGS="-j2"
#-- Debugging flags
DEBUG_CFLAGS="-g -fvar-tracking-assignments"
DEBUG_CXXFLAGS="-g -fvar-tracking-assignments"
#DEBUG_RUSTFLAGS="-C debuginfo=2"

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[dmalcolm at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #12 from David Malcolm <dmalcolm at gcc dot gnu.org> ---
Is that the default /etc/makepkg.conf, or did you hand-edit it?  (i.e is this
something that all AUR users are going to run into, or just you?)

Clearly the "-Werror=format-security" is not compatible with building
libgccjit; I think "-fexceptions" is probably not a good idea either.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #13 from sujay1844 at protonmail dot com ---
(In reply to David Malcolm from comment #12)
> Is that the default /etc/makepkg.conf, or did you hand-edit it?  (i.e is
> this something that all AUR users are going to run into, or just you?)
> 
> Clearly the "-Werror=format-security" is not compatible with building
> libgccjit; I think "-fexceptions" is probably not a good idea either.

Nope I don't think I ever opened this. In fact, I just came to know that it
exists a few minutes back. Please suggest the settings to change.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[dmalcolm at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #14 from David Malcolm <dmalcolm at gcc dot gnu.org> ---
(In reply to sujay1844 from comment #13)
> (In reply to David Malcolm from comment #12)
> > Is that the default /etc/makepkg.conf, or did you hand-edit it?  (i.e is
> > this something that all AUR users are going to run into, or just you?)
> > 
> > Clearly the "-Werror=format-security" is not compatible with building
> > libgccjit; I think "-fexceptions" is probably not a good idea either.
> 
> Nope I don't think I ever opened this. In fact, I just came to know that it
> exists a few minutes back. Please suggest the settings to change.

I wonder if it's just you that's running into this, or it's affecting everyone
using libgccjit via AUR.

I think you need to take this downstream, to the AUR community; it appears that
the default /etc/makepkg.conf is injecting build flags that are incompatible
with libgccjit, so either the default needs to change, or libgccjit's PKGBUILD
needs to change.  Hopefully the above discussion has enough information for the
downstream AUR experts to fix things.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #15 from sujay1844 at protonmail dot com ---
I did some digging and found this:-

I checked my timeshift snapshot (which I took just after installation). The
etc/makepkg.conf is that goes like so:-

#-- Compiler and Linker Flags
CPPFLAGS="-D_FORTIFY_SOURCE=2"
CFLAGS="-march=x86-64 -mtune=generic -O2 -pipe -fno-plt"
CXXFLAGS="-march=x86-64 -mtune=generic -O2 -pipe -fno-plt"
LDFLAGS="-Wl,-O1,--sort-common,--as-needed,-z,relro,-z,now"
#RUSTFLAGS="-C opt-level=2"
#-- Make Flags: change this for DistCC/SMP systems
#MAKEFLAGS="-j2"
#-- Debugging flags
DEBUG_CFLAGS="-g -fvar-tracking-assignments"
DEBUG_CXXFLAGS="-g -fvar-tracking-assignments"
#DEBUG_RUSTFLAGS="-C debuginfo=2"


To be sure that this is safe, I installed Manjaro in a virtual machine.
(Clearly, the guys at Manjaro know more about security than me) The
/etc/makepkg.conf had the same thing as in my timeshift snapshot. So I conclude
that removing those unwanted flags are safe. I also checked the ArchWiki
article on makepkg and made some tweaks to improve performance and this is my
current /etc/makepkg.conf looks like:-

#-- Compiler and Linker Flags
CPPFLAGS="-D_FORTIFY_SOURCE=2"
CFLAGS="-march=x86-64 -mtune=generic -O2 -pipe -fno-plt"
CXXFLAGS="-march=x86-64 -mtune=generic -O2 -pipe -fno-plt"
LDFLAGS="-Wl,-O1,--sort-common,--as-needed,-z,relro,-z,now"
#RUSTFLAGS="-C opt-level=2"
#-- Make Flags: change this for DistCC/SMP systems
#MAKEFLAGS="-j2"
#-- Debugging flags
DEBUG_CFLAGS="-g -fvar-tracking-assignments"
DEBUG_CXXFLAGS="-g -fvar-tracking-assignments"
#DEBUG_RUSTFLAGS="-C debuginfo=2"


Thank you for your valuable time!!
Please give any suggestions if any.
And you can close this bug.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[jonathon at m2x dot dev]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

Jonathon <jonathon at m2x dot dev> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jonathon at m2x dot dev

--- Comment #16 from Jonathon <jonathon at m2x dot dev> ---
> Kernel: 5.11.15-arch1-2

> the guys at Manjaro

If you are running Manjaro please don't refer to Arch. They are not the same
distribution and it causes huge amounts of confusion given the large number of
patches and changes that Manjaro apply.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #17 from sujay1844 at protonmail dot com ---
No duh I'm not running Manjaro.I said I installed Manjaro in a virtual machine
to check what the guys at Manjaro did to the /etc/makepkg.conf. Please read the
context before posting such comments.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[jonathon at m2x dot dev]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #18 from Jonathon <jonathon at m2x dot dev> ---
Thanks for clearing that up. You gave the impression that you reinstalled
Manjaro in a VM to ensure it wasn't a local change, and it wasn't otherwise
clear what Manjaro had to do with anything (i.e. why would Manjaro's settings
indicate what was "safe" in Arch?).

As it is, have a look at the Arch bug report and see if there's anything you
can add: https://bugs.archlinux.org/task/70701

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #19 from sujay1844 at protonmail dot com ---
Because the guys at Manjaro are much more knowledgeable than me to know what is
safe and what is not IMO.

[Bug jit/100207] -Werror=format-security error in AUR build of libgccjit

[sujay1844 at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100207

--- Comment #20 from sujay1844 at protonmail dot com ---
Dude is it right to file a bug report?? Because I'm not sure if that was the
default in Arch. Maybe some stupid program messed it up??