linux-kernel - Re: arch/riscv/net/bpf_jit.h:241:30: sparse: sparse: cast truncates bits from constant value (7ff becomes ff)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <3d0797499c08d367f69487f95ae884f7b5bfdfcb.camel@intel.com>
Date:   Mon, 31 Jul 2023 09:38:44 +0000
From:   "Liu, Yujie" <yujie.liu@...el.com>
To:     "rdunlap@...radead.org" <rdunlap@...radead.org>
CC:     "daniel@...earbox.net" <daniel@...earbox.net>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "oe-kbuild-all@...ts.linux.dev" <oe-kbuild-all@...ts.linux.dev>,
        lkp <lkp@...el.com>,
        "lukenels@...washington.edu" <lukenels@...washington.edu>
Subject: Re: arch/riscv/net/bpf_jit.h:241:30: sparse: sparse: cast truncates
 bits from constant value (7ff becomes ff)

Hi Randy,

On Tue, 2023-07-25 at 21:26 -0700, Randy Dunlap wrote:
> [add Luke Nelson]
> 
> [why on earth was this sent to me?]

We investigated this case and found that it was a false positive and
not caused by your commit. Sorry for that.

This report was generated when running sparse. We used to catch a build
error as below:

  CC      arch/riscv/net/bpf_jit_comp64.o
arch/riscv/net/bpf_jit_comp64.c: In function 'bpf_arch_text_poke':
arch/riscv/net/bpf_jit_comp64.c:691:23: error: implicit declaration of
function 'patch_text'; did you mean 'path_get'? [-Werror=implicit-
function-declaration]
  691 |                 ret = patch_text(ip, new_insns, ninsns);
      |                       ^~~~~~~~~~
      |                       path_get

This is exactly what commit 2d311f480b52 wants to fix. Then the build
error is gone and a sparse warning jumps out as below:

  CC      arch/riscv/net/bpf_jit_comp64.o  <-- compiling stage
  CHECK   arch/riscv/net/bpf_jit_comp64.c  <-- sparse check stage
WARNING: invalid argument to '-march': '_zicbom_zihintpause'
arch/riscv/net/bpf_jit_comp64.c: note: in included file:
arch/riscv/net/bpf_jit.h:241:30: sparse: sparse: cast truncates bits
from constant value (7ff becomes ff)

Our bisection logic wrongly assumed that a new issue was introduced on
this commit, but actually it wasn't. It should be an already-there
problem in existing code.

Sorry again for any inconvenience. We will filter out this false
positive and optimize the bot.

Best Regards,
Yujie

> On 7/25/23 16:34, kernel test robot wrote:
> > tree:  
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 
> > master
> > head:   af2e19389c2c1d8a299e04c5105b180ef0c5b5b5
> > commit: 2d311f480b52eeb2e1fd432d64b78d82952c3808 riscv, bpf: Fix
> > patch_text implicit declaration
> > date:   5 months ago
> > config: riscv-randconfig-r072-20230725
> > (https://download.01.org/0day-ci/archive/20230726/202307260704.dUEl
> > CrWU-lkp@...el.com/config)
> > compiler: riscv64-linux-gcc (GCC) 12.3.0
> > reproduce:
> > (https://download.01.org/0day-ci/archive/20230726/202307260704.dUEl
> > CrWU-lkp@...el.com/reproduce)
> > 
> > If you fix the issue in a separate patch/commit (i.e. not just a
> > new version of
> > the same patch/commit), kindly add following tags
> > > Reported-by: kernel test robot <lkp@...el.com>
> > > Closes:
> > > https://lore.kernel.org/oe-kbuild-all/202307260704.dUElCrWU-lkp@intel.com/
> > 
> > sparse warnings: (new ones prefixed by >>)
> >    WARNING: invalid argument to '-march': '_zicbom_zihintpause'
> >    arch/riscv/net/bpf_jit_comp64.c: note: in included file:
> > > > arch/riscv/net/bpf_jit.h:241:30: sparse: sparse: cast truncates
> > > > bits from constant value (7ff becomes ff)
> > 
> > vim +241 arch/riscv/net/bpf_jit.h
> > 
> > ca6cb5447ceca6 Luke Nelson 2020-03-04  238  
> > ca6cb5447ceca6 Luke Nelson 2020-03-04  239  static inline u32
> > rv_s_insn(u16 imm11_0, u8 rs2, u8 rs1, u8 funct3, u8 opcode)
> > ca6cb5447ceca6 Luke Nelson 2020-03-04  240  {
> > ca6cb5447ceca6 Luke Nelson 2020-03-04 @241      u8 imm11_5 =
> > imm11_0 >> 5, imm4_0 = imm11_0 & 0x1f;
> > ca6cb5447ceca6 Luke Nelson 2020-03-04  242  
> > ca6cb5447ceca6 Luke Nelson 2020-03-04  243      return (imm11_5 <<
> > 25) | (rs2 << 20) | (rs1 << 15) | (funct3 << 12) |
> > ca6cb5447ceca6 Luke Nelson 2020-03-04  244              (imm4_0 <<
> > 7) | opcode;
> > ca6cb5447ceca6 Luke Nelson 2020-03-04  245  }
> > ca6cb5447ceca6 Luke Nelson 2020-03-04  246  
> > 
> > :::::: The code at line 241 was first introduced by commit
> > :::::: ca6cb5447ceca6a87d6b62c9e5d41042c34f7ffa riscv, bpf: Factor
> > common RISC-V JIT code
> > 
> > :::::: TO: Luke Nelson <lukenels@...washington.edu>
> > :::::: CC: Daniel Borkmann <daniel@...earbox.net>
> > 
>