linux-kernel - Re: [PATCH -next] iwlwifi: pcie: Use bitwise instead of arithmetic operator for flags

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 for Android: free password hash cracker in your pocket

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

Date:   Wed, 06 May 2020 08:15:08 -0700
From:   Joe Perches <joe@...ches.com>
To:     Luciano Coelho <luciano.coelho@...el.com>,
        Samuel Zou <zou_wei@...wei.com>, johannes.berg@...el.com,
        emmanuel.grumbach@...el.com, linuxwifi@...el.com,
        kvalo@...eaurora.org, davem@...emloft.net,
        Julia Lawall <julia.lawall@...6.fr>
Cc:     linux-wireless@...r.kernel.org, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org, cocci <cocci@...teme.lip6.fr>
Subject: Re: [PATCH -next] iwlwifi: pcie: Use bitwise instead of arithmetic
 operator for flags

On Wed, 2020-05-06 at 16:51 +0300, Luciano Coelho wrote:
> On Tue, 2020-05-05 at 20:19 -0700, Joe Perches wrote:
> > On Wed, 2020-05-06 at 11:07 +0800, Samuel Zou wrote:
> > > This silences the following coccinelle warning:
> > > 
> > > "WARNING: sum of probable bitmasks, consider |"
> > 
> > I suggest instead ignoring bad and irrelevant warnings.
> > 
> > PREFIX_LEN is 32 not 0x20 or BIT(5)
> > PCI_DUMP_SIZE is 352
> > 
> > > diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
> > []
> > > @@ -109,9 +109,9 @@ void iwl_trans_pcie_dump_regs(struct iwl_trans *trans)
> > >  
> > >  	/* Alloc a max size buffer */
> > >  	alloc_size = PCI_ERR_ROOT_ERR_SRC +  4 + PREFIX_LEN;
> > > -	alloc_size = max_t(u32, alloc_size, PCI_DUMP_SIZE + PREFIX_LEN);
> > > -	alloc_size = max_t(u32, alloc_size, PCI_MEM_DUMP_SIZE + PREFIX_LEN);
> > > -	alloc_size = max_t(u32, alloc_size, PCI_PARENT_DUMP_SIZE + PREFIX_LEN);
> > > +	alloc_size = max_t(u32, alloc_size, PCI_DUMP_SIZE | PREFIX_LEN);
> > > +	alloc_size = max_t(u32, alloc_size, PCI_MEM_DUMP_SIZE | PREFIX_LEN);
> > > +	alloc_size = max_t(u32, alloc_size, PCI_PARENT_DUMP_SIZE | PREFIX_LEN);
> > >  
> > >  	buf = kmalloc(alloc_size, GFP_ATOMIC);
> > >  	if (!buf)
> 
> Yeah, those macros are clearly not bitmasks.  I'm dropping this patch.

Can the cocci script that generated this warning

scripts/coccinelle/misc/orplus.cocci

be dropped or improved to validate the likelihood that
the defines or constants used are more likely than
not are bit values?

Maybe these should be defined as hex or BIT or BIT_ULL
or GENMASK or the like?


Right now it seems it just tests for two constants.