Re: [PATCH RFC 5/5] non-mm: discourage the usage of __GFP_NOFAIL and encourage GFP_NOFAIL

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed 24-07-24 06:13:31, Christoph Hellwig wrote:
> On Wed, Jul 24, 2024 at 11:53:49AM +0200, Vlastimil Babka wrote:
> > GFP_KERNEL_NOFAIL which is GFP_KERNEL | __GFP_NOFAIL
> > 
> > And probably also GFP_NOFS_NOFAIL and GFP_NOIO_NOFAIL (sigh).
> 
> Let's not add these and force people to use the scope API which
> we're trying to move to.  I think we should be able to simply
> have GFP_NOFAIL which includes GFP_KERNEL and no variant.

Scope API is tricky here. Exactly because the scope itself could have
opportunistic GFP_NOWAIT allocations. It just takes a library function
to call from a scope that could become a risk because it doesn't know
about potential scopes it can live in.

-- 
Michal Hocko
SUSE Labs




[Index of Archives]     [KVM Development]     [Libvirt Development]     [Libvirt Users]     [CentOS Virtualization]     [Netdev]     [Ethernet Bridging]     [Linux Wireless]     [Kernel Newbies]     [Security]     [Linux for Hams]     [Netfilter]     [Bugtraq]     [Yosemite Forum]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux Admin]     [Samba]

  Powered by Linux