From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.linuxfoundation.org ([140.211.169.12]:46414 "EHLO mail.linuxfoundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753006AbbIQFZ5 (ORCPT ); Thu, 17 Sep 2015 01:25:57 -0400 Subject: Patch "crypto: vmx - Fixing AES-CTR counter bug" has been added to the 4.2-stable tree To: leosilva@linux.vnet.ibm.com, gregkh@linuxfoundation.org, herbert@gondor.apana.org.au Cc: , From: Date: Wed, 16 Sep 2015 22:25:47 -0700 Message-ID: <14424675471297@kroah.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ANSI_X3.4-1968 Content-Transfer-Encoding: 8bit Sender: stable-owner@vger.kernel.org List-ID: This is a note to let you know that I've just added the patch titled crypto: vmx - Fixing AES-CTR counter bug to the 4.2-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: crypto-vmx-fixing-aes-ctr-counter-bug.patch and it can be found in the queue-4.2 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let know about it. >>From 1d4aa0b4c1816e8ca92a6aadb0d8f6b43c56c0d0 Mon Sep 17 00:00:00 2001 From: Leonidas Da Silva Barbosa Date: Fri, 14 Aug 2015 10:12:22 -0300 Subject: crypto: vmx - Fixing AES-CTR counter bug From: Leonidas Da Silva Barbosa commit 1d4aa0b4c1816e8ca92a6aadb0d8f6b43c56c0d0 upstream. AES-CTR is using a counter 8bytes-8bytes what miss match with kernel specs. In the previous code a vadduwm was done to increment counter. Replacing this for a vadduqm now considering both cases counter 8-8 bytes and full 16bytes. Signed-off-by: Leonidas S Barbosa Signed-off-by: Herbert Xu Signed-off-by: Greg Kroah-Hartman --- drivers/crypto/vmx/aes_ctr.c | 8 +++++++- drivers/crypto/vmx/aesp8-ppc.pl | 34 +++++++++++++++++----------------- 2 files changed, 24 insertions(+), 18 deletions(-) --- a/drivers/crypto/vmx/aes_ctr.c +++ b/drivers/crypto/vmx/aes_ctr.c @@ -113,6 +113,7 @@ static int p8_aes_ctr_crypt(struct blkci struct scatterlist *src, unsigned int nbytes) { int ret; + u64 inc; struct blkcipher_walk walk; struct p8_aes_ctr_ctx *ctx = crypto_tfm_ctx(crypto_blkcipher_tfm(desc->tfm)); @@ -140,7 +141,12 @@ static int p8_aes_ctr_crypt(struct blkci walk.iv); pagefault_enable(); - crypto_inc(walk.iv, AES_BLOCK_SIZE); + /* We need to update IV mostly for last bytes/round */ + inc = (nbytes & AES_BLOCK_MASK) / AES_BLOCK_SIZE; + if (inc > 0) + while (inc--) + crypto_inc(walk.iv, AES_BLOCK_SIZE); + nbytes &= AES_BLOCK_SIZE - 1; ret = blkcipher_walk_done(desc, &walk, nbytes); } --- a/drivers/crypto/vmx/aesp8-ppc.pl +++ b/drivers/crypto/vmx/aesp8-ppc.pl @@ -1437,28 +1437,28 @@ Load_ctr32_enc_key: ?vperm v31,v31,$out0,$keyperm lvx v25,$x10,$key_ # pre-load round[2] - vadduwm $two,$one,$one + vadduqm $two,$one,$one subi $inp,$inp,15 # undo "caller" $SHL $len,$len,4 - vadduwm $out1,$ivec,$one # counter values ... - vadduwm $out2,$ivec,$two + vadduqm $out1,$ivec,$one # counter values ... + vadduqm $out2,$ivec,$two vxor $out0,$ivec,$rndkey0 # ... xored with rndkey[0] le?li $idx,8 - vadduwm $out3,$out1,$two + vadduqm $out3,$out1,$two vxor $out1,$out1,$rndkey0 le?lvsl $inpperm,0,$idx - vadduwm $out4,$out2,$two + vadduqm $out4,$out2,$two vxor $out2,$out2,$rndkey0 le?vspltisb $tmp,0x0f - vadduwm $out5,$out3,$two + vadduqm $out5,$out3,$two vxor $out3,$out3,$rndkey0 le?vxor $inpperm,$inpperm,$tmp # transform for lvx_u/stvx_u - vadduwm $out6,$out4,$two + vadduqm $out6,$out4,$two vxor $out4,$out4,$rndkey0 - vadduwm $out7,$out5,$two + vadduqm $out7,$out5,$two vxor $out5,$out5,$rndkey0 - vadduwm $ivec,$out6,$two # next counter value + vadduqm $ivec,$out6,$two # next counter value vxor $out6,$out6,$rndkey0 vxor $out7,$out7,$rndkey0 @@ -1594,27 +1594,27 @@ Loop_ctr32_enc8x_middle: vcipherlast $in0,$out0,$in0 vcipherlast $in1,$out1,$in1 - vadduwm $out1,$ivec,$one # counter values ... + vadduqm $out1,$ivec,$one # counter values ... vcipherlast $in2,$out2,$in2 - vadduwm $out2,$ivec,$two + vadduqm $out2,$ivec,$two vxor $out0,$ivec,$rndkey0 # ... xored with rndkey[0] vcipherlast $in3,$out3,$in3 - vadduwm $out3,$out1,$two + vadduqm $out3,$out1,$two vxor $out1,$out1,$rndkey0 vcipherlast $in4,$out4,$in4 - vadduwm $out4,$out2,$two + vadduqm $out4,$out2,$two vxor $out2,$out2,$rndkey0 vcipherlast $in5,$out5,$in5 - vadduwm $out5,$out3,$two + vadduqm $out5,$out3,$two vxor $out3,$out3,$rndkey0 vcipherlast $in6,$out6,$in6 - vadduwm $out6,$out4,$two + vadduqm $out6,$out4,$two vxor $out4,$out4,$rndkey0 vcipherlast $in7,$out7,$in7 - vadduwm $out7,$out5,$two + vadduqm $out7,$out5,$two vxor $out5,$out5,$rndkey0 le?vperm $in0,$in0,$in0,$inpperm - vadduwm $ivec,$out6,$two # next counter value + vadduqm $ivec,$out6,$two # next counter value vxor $out6,$out6,$rndkey0 le?vperm $in1,$in1,$in1,$inpperm vxor $out7,$out7,$rndkey0 Patches currently in stable-queue which might be from leosilva@linux.vnet.ibm.com are queue-4.2/crypto-vmx-fixing-ghash-key-issue-on-little-endian.patch queue-4.2/crypto-vmx-fixing-aes-ctr-counter-bug.patch