Revision 0ee931c4e31a5efb134c76440405e9219f896e33 authored by Michal Hocko on 13 September 2017, 23:28:29 UTC, committed by Linus Torvalds on 14 September 2017, 01:53:16 UTC
GFP_TEMPORARY was introduced by commit e12ba74d8ff3 ("Group short-lived
and reclaimable kernel allocations") along with __GFP_RECLAIMABLE.  It's
primary motivation was to allow users to tell that an allocation is
short lived and so the allocator can try to place such allocations close
together and prevent long term fragmentation.  As much as this sounds
like a reasonable semantic it becomes much less clear when to use the
highlevel GFP_TEMPORARY allocation flag.  How long is temporary? Can the
context holding that memory sleep? Can it take locks? It seems there is
no good answer for those questions.

The current implementation of GFP_TEMPORARY is basically GFP_KERNEL |
__GFP_RECLAIMABLE which in itself is tricky because basically none of
the existing caller provide a way to reclaim the allocated memory.  So
this is rather misleading and hard to evaluate for any benefits.

I have checked some random users and none of them has added the flag
with a specific justification.  I suspect most of them just copied from
other existing users and others just thought it might be a good idea to
use without any measuring.  This suggests that GFP_TEMPORARY just
motivates for cargo cult usage without any reasoning.

I believe that our gfp flags are quite complex already and especially
those with highlevel semantic should be clearly defined to prevent from
confusion and abuse.  Therefore I propose dropping GFP_TEMPORARY and
replace all existing users to simply use GFP_KERNEL.  Please note that
SLAB users with shrinkers will still get __GFP_RECLAIMABLE heuristic and
so they will be placed properly for memory fragmentation prevention.

I can see reasons we might want some gfp flag to reflect shorterm
allocations but I propose starting from a clear semantic definition and
only then add users with proper justification.

This was been brought up before LSF this year by Matthew [1] and it
turned out that GFP_TEMPORARY really doesn't have a clear semantic.  It
seems to be a heuristic without any measured advantage for most (if not
all) its current users.  The follow up discussion has revealed that
opinions on what might be temporary allocation differ a lot between
developers.  So rather than trying to tweak existing users into a
semantic which they haven't expected I propose to simply remove the flag
and start from scratch if we really need a semantic for short term
allocations.

[1] http://lkml.kernel.org/r/20170118054945.GD18349@bombadil.infradead.org

[akpm@linux-foundation.org: fix typo]
[akpm@linux-foundation.org: coding-style fixes]
[sfr@canb.auug.org.au: drm/i915: fix up]
  Link: http://lkml.kernel.org/r/20170816144703.378d4f4d@canb.auug.org.au
Link: http://lkml.kernel.org/r/20170728091904.14627-1-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Acked-by: Mel Gorman <mgorman@suse.de>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Neil Brown <neilb@suse.de>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent d0dbf77
Raw File
extract-module-sig.pl
#!/usr/bin/env perl
#
# extract-mod-sig <part> <module-file>
#
# Reads the module file and writes out some or all of the signature
# section to stdout.  Part is the bit to be written and is one of:
#
#  -0: The unsigned module, no signature data at all
#  -a: All of the signature data, including magic number
#  -d: Just the descriptor values as a sequence of numbers
#  -n: Just the signer's name
#  -k: Just the key ID
#  -s: Just the crypto signature or PKCS#7 message
#
use warnings;
use strict;

die "Format: $0 -[0adnks] module-file >out\n"
    if ($#ARGV != 1);

my $part = $ARGV[0];
my $modfile = $ARGV[1];

my $magic_number = "~Module signature appended~\n";

#
# Read the module contents
#
open FD, "<$modfile" || die $modfile;
binmode(FD);
my @st = stat(FD);
die "$modfile" unless (@st);
my $buf = "";
my $len = sysread(FD, $buf, $st[7]);
die "$modfile" unless (defined($len));
die "Short read on $modfile\n" unless ($len == $st[7]);
close(FD) || die $modfile;

print STDERR "Read ", $len, " bytes from module file\n";

die "The file is too short to have a sig magic number and descriptor\n"
    if ($len < 12 + length($magic_number));

#
# Check for the magic number and extract the information block
#
my $p = $len - length($magic_number);
my $raw_magic = substr($buf, $p);

die "Magic number not found at $len\n"
    if ($raw_magic ne $magic_number);
print STDERR "Found magic number at $len\n";

$p -= 12;
my $raw_info = substr($buf, $p, 12);

my @info = unpack("CCCCCxxxN", $raw_info);
my ($algo, $hash, $id_type, $name_len, $kid_len, $sig_len) = @info;

if ($id_type == 0) {
    print STDERR "Found PGP key identifier\n";
} elsif ($id_type == 1) {
    print STDERR "Found X.509 cert identifier\n";
} elsif ($id_type == 2) {
    print STDERR "Found PKCS#7/CMS encapsulation\n";
} else {
    print STDERR "Found unsupported identifier type $id_type\n";
}

#
# Extract the three pieces of info data
#
die "Insufficient name+kid+sig data in file\n"
    unless ($p >= $name_len + $kid_len + $sig_len);

$p -= $sig_len;
my $raw_sig = substr($buf, $p, $sig_len);
$p -= $kid_len;
my $raw_kid = substr($buf, $p, $kid_len);
$p -= $name_len;
my $raw_name = substr($buf, $p, $name_len);

my $module_len = $p;

if ($sig_len > 0) {
    print STDERR "Found $sig_len bytes of signature [";
    my $n = $sig_len > 16 ? 16 : $sig_len;
    foreach my $i (unpack("C" x $n, substr($raw_sig, 0, $n))) {
	printf STDERR "%02x", $i;
    }
    print STDERR "]\n";
}

if ($kid_len > 0) {
    print STDERR "Found $kid_len bytes of key identifier [";
    my $n = $kid_len > 16 ? 16 : $kid_len;
    foreach my $i (unpack("C" x $n, substr($raw_kid, 0, $n))) {
	printf STDERR "%02x", $i;
    }
    print STDERR "]\n";
}

if ($name_len > 0) {
    print STDERR "Found $name_len bytes of signer's name [$raw_name]\n";
}

#
# Produce the requested output
#
if ($part eq "-0") {
    # The unsigned module, no signature data at all
    binmode(STDOUT);
    print substr($buf, 0, $module_len);
} elsif ($part eq "-a") {
    # All of the signature data, including magic number
    binmode(STDOUT);
    print substr($buf, $module_len);
} elsif ($part eq "-d") {
    # Just the descriptor values as a sequence of numbers
    print join(" ", @info), "\n";
} elsif ($part eq "-n") {
    # Just the signer's name
    print STDERR "No signer's name for PKCS#7 message type sig\n"
	if ($id_type == 2);
    binmode(STDOUT);
    print $raw_name;
} elsif ($part eq "-k") {
    # Just the key identifier
    print STDERR "No key ID for PKCS#7 message type sig\n"
	if ($id_type == 2);
    binmode(STDOUT);
    print $raw_kid;
} elsif ($part eq "-s") {
    # Just the crypto signature or PKCS#7 message
    binmode(STDOUT);
    print $raw_sig;
}
back to top