[PATCHv2] KVM: unbreak userspace that does not sets tss address

2011-03-13 Thread Gleb Natapov
Commit 6440e5967bc broke old userspaces that do not set tss address before entering vcpu. Unbreak it by setting tss address to a safe value on the first vcpu entry. New userspaces should set tss address, so print warning in case it doesn't. Signed-off-by: Gleb Natapov g...@redhat.com --- v1-v2

Re: [PATCH 3/3] vhost-net: use lock_sock_fast() in peek_head_len()

2011-03-13 Thread Michael S. Tsirkin
On Mon, Jan 17, 2011 at 04:11:17PM +0800, Jason Wang wrote: We can use lock_sock_fast() instead of lock_sock() in order to get speedup in peek_head_len(). Signed-off-by: Jason Wang jasow...@redhat.com --- drivers/vhost/net.c |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)

[PATCH 0/5] KVM: x86 emulator: Use single stage decoding -- Part 1

2011-03-13 Thread Takuya Yoshikawa
This work will continue until we can remove the ugly switch statements. But I want to do this with enough care not to insert extra errors. -- For me, this is a good opportunity to read SDM well. So the whole work will be done in a step by step manner! Thanks, Takuya -- To unsubscribe from

[PATCH 1/5] KVM: x86 emulator: Use single stage decoding for Group 1 instructions

2011-03-13 Thread Takuya Yoshikawa
From: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp ADD, OR, ADC, SBB, AND, SUB, XOR, CMP are converted using a new macro I6ALU(_f, _e). CMPS, SCAS will be converted later. Signed-off-by: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp --- arch/x86/kvm/emulate.c | 151

[PATCH 2/5] KVM: x86 emulator: Use single stage decoding for PUSH/POP XS instructions

2011-03-13 Thread Takuya Yoshikawa
From: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp PUSH ES/CS/SS/DS/FS/GS and POP ES/SS/DS/FS/GS are converted. Signed-off-by: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp --- arch/x86/kvm/emulate.c | 111 +++- 1 files changed, 72

[PATCH 3/5] KVM: x86 emulator: Use single stage decoding for POP instructions

2011-03-13 Thread Takuya Yoshikawa
From: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp POP is converted. RET will be converted later. Signed-off-by: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp --- arch/x86/kvm/emulate.c | 16 ++-- 1 files changed, 10 insertions(+), 6 deletions(-) diff --git

[PATCH 4/5] KVM: x86 emulator: Use single stage decoding for PUSHA and POPA instructions

2011-03-13 Thread Takuya Yoshikawa
From: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp PUSHA and POPA are converted. Signed-off-by: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp --- arch/x86/kvm/emulate.c | 19 --- 1 files changed, 12 insertions(+), 7 deletions(-) diff --git a/arch/x86/kvm/emulate.c

[PATCH 5/5] KVM: x86 emulator: Use single stage decoding for PUSHF and POPF instructions

2011-03-13 Thread Takuya Yoshikawa
From: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp PUSHF and POPF are converted. Signed-off-by: Takuya Yoshikawa yoshikawa.tak...@oss.ntt.co.jp --- arch/x86/kvm/emulate.c | 32 +--- 1 files changed, 21 insertions(+), 11 deletions(-) diff --git

Re: [PATCH 3/3] vhost-net: use lock_sock_fast() in peek_head_len()

2011-03-13 Thread Eric Dumazet
Le dimanche 13 mars 2011 à 17:06 +0200, Michael S. Tsirkin a écrit : On Mon, Jan 17, 2011 at 04:11:17PM +0800, Jason Wang wrote: We can use lock_sock_fast() instead of lock_sock() in order to get speedup in peek_head_len(). Signed-off-by: Jason Wang jasow...@redhat.com ---

Re: [PATCH 3/3] vhost-net: use lock_sock_fast() in peek_head_len()

2011-03-13 Thread Michael S. Tsirkin
On Sun, Mar 13, 2011 at 04:52:50PM +0100, Eric Dumazet wrote: Le dimanche 13 mars 2011 à 17:06 +0200, Michael S. Tsirkin a écrit : On Mon, Jan 17, 2011 at 04:11:17PM +0800, Jason Wang wrote: We can use lock_sock_fast() instead of lock_sock() in order to get speedup in peek_head_len().

Re: [PATCH 3/3] vhost-net: use lock_sock_fast() in peek_head_len()

2011-03-13 Thread Eric Dumazet
Le dimanche 13 mars 2011 à 18:19 +0200, Michael S. Tsirkin a écrit : Other side is in drivers/net/tun.c and net/packet/af_packet.c At least wrt tun it seems clear socket is not locked. Yes (assuming you refer to tun_net_xmit()) Besides queue, dequeue seems to be done without socket locked.

Re: [PATCH 3/3] vhost-net: use lock_sock_fast() in peek_head_len()

2011-03-13 Thread Michael S. Tsirkin
On Sun, Mar 13, 2011 at 05:32:07PM +0100, Eric Dumazet wrote: Le dimanche 13 mars 2011 à 18:19 +0200, Michael S. Tsirkin a écrit : Other side is in drivers/net/tun.c and net/packet/af_packet.c At least wrt tun it seems clear socket is not locked. Yes (assuming you refer to

Re: [PATCH 3/3] vhost-net: use lock_sock_fast() in peek_head_len()

2011-03-13 Thread Eric Dumazet
Le dimanche 13 mars 2011 à 18:43 +0200, Michael S. Tsirkin a écrit : On Sun, Mar 13, 2011 at 05:32:07PM +0100, Eric Dumazet wrote: Le dimanche 13 mars 2011 à 18:19 +0200, Michael S. Tsirkin a écrit : Other side is in drivers/net/tun.c and net/packet/af_packet.c At least wrt tun it

Re: [PATCH] KVM Test: nicdriver_unload.py: Fix some failure whan running RHEL4.8 guest

2011-03-13 Thread Lucas Meneghel Rodrigues
On Fri, 2011-03-04 at 14:14 +0800, Qingtang Zhou wrote: 1. readlink in RHEL4.8 doesn't have a '-e' parameter, replace it with '-f'. 2. nic driver parameter path in '/sys' is different, get it from config file. Hi, thanks for your patch! A little comment below: Signed-off-by: Qingtang Zhou

Re: [PATCH 3/3] vhost-net: use lock_sock_fast() in peek_head_len()

2011-03-13 Thread Michael S. Tsirkin
On Sun, Mar 13, 2011 at 06:41:32PM +0100, Eric Dumazet wrote: Le dimanche 13 mars 2011 à 18:43 +0200, Michael S. Tsirkin a écrit : On Sun, Mar 13, 2011 at 05:32:07PM +0100, Eric Dumazet wrote: Le dimanche 13 mars 2011 à 18:19 +0200, Michael S. Tsirkin a écrit : Other side is in