ShuriZma
/
xemu
mirror of https://github.com/xemu-project/xemu.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
xemu/hw
History
yuchenlin 157628d067 e1000: Delay flush queue when receive RCTL 
Due to too early RCT0 interrput, win10x32 may hang on booting.
This problem can be reproduced by doing power cycle on win10x32 guest.
In our environment, we have 10 win10x32 and stress power cycle.
The problem will happen about 20 rounds.

Below shows some log with comment:

The normal case:

22831@1551928392.984687:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
22831@1551928392.985655:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
22831@1551928392.985801:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
22831@1551928393.056710:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: ICR read: 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
22831@1551928393.077548:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: ICR read: 0
e1000: set_ics 2, ICR 0, IMR 0
e1000: set_ics 2, ICR 2, IMR 0
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
22831@1551928393.102974:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
22831@1551928393.103267:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
e1000: RCTL: 255, mac_reg[RCTL] = 0x40002 <- win10x32 says it can handle
RX now
e1000: set_ics 0, ICR 2, IMR 9d <- unmask interrupt
e1000: RCTL: 255, mac_reg[RCTL] = 0x48002
e1000: set_ics 80, ICR 2, IMR 9d <- interrupt and work!
...

The bad case:

27744@1551930483.117766:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
27744@1551930483.118398:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
27744@1551930483.198063:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: ICR read: 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
27744@1551930483.218675:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
e1000: set_ics 0, ICR 0, IMR 0
e1000: ICR read: 0
e1000: set_ics 2, ICR 0, IMR 0
e1000: set_ics 2, ICR 2, IMR 0
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
27744@1551930483.241768:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
27744@1551930483.241979:e1000x_rx_disabled Received packet dropped
because receive is disabled RCTL = 0
e1000: RCTL: 255, mac_reg[RCTL] = 0x40002 <- win10x32 says it can handle
RX now
e1000: set_ics 80, ICR 2, IMR 0 <- flush queue (caused by setting RCTL)
e1000: set_ics 0, ICR 82, IMR 9d <- unmask interrupt and because 0x82&0x9d
!= 0 generate interrupt, hang on here...

To workaround this problem, simply delay flush queue. Also stop receiving
when timer is going to run.

Tested on CentOS, Win7SP1x64 and Win10x32.

Signed-off-by: yuchenlin <yuchenlin@synology.com>
Reviewed-by: Dmitry Fleytman <dmitry.fleytman@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
 		2019-03-29 15:22:18 +08:00
..
9pfs trace-events: Fix attribution of trace points to source 2019-03-22 16:18:07 +00:00
acpi trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
adc kconfig: introduce kconfig files 2019-03-07 21:45:53 +01:00
alpha * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
arm trace-events: Fix attribution of trace points to source 2019-03-22 16:18:07 +00:00
audio trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
block Pflash and firmware configuration patches for 2019-03-26 2019-03-26 09:57:07 +00:00
bt kconfig: introduce kconfig files 2019-03-07 21:45:53 +01:00
char * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
core Machine queue, 2019-03-11 2019-03-12 15:25:46 +00:00
cpu kconfig: introduce kconfig files 2019-03-07 21:45:53 +01:00
cris cris-softmmu.mak: express dependencies with Kconfig 2019-03-07 21:46:19 +01:00
display * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
dma trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
gpio Pull request 2019-03-25 17:01:10 +00:00
hppa * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
hyperv hyperv: express dependencies with kconfig 2019-03-07 21:45:53 +01:00
i2c trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
i386 * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
ide trace-events: Fix attribution of trace points to source 2019-03-22 16:18:07 +00:00
input trace-events: Fix attribution of trace points to source 2019-03-22 16:18:07 +00:00
intc * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
ipack build: convert pci.mak to Kconfig 2019-03-07 21:45:53 +01:00
ipmi ipmi: express dependencies with kconfig 2019-03-07 21:45:53 +01:00
isa * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
lm32 pflash: Clean up after commit 368a354f02, part 2 2019-03-11 22:53:44 +01:00
m68k m68k-softmmu.mak: express dependencies with Kconfig 2019-03-07 21:46:19 +01:00
mem trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
microblaze pflash: Clean up after commit 368a354f02, part 2 2019-03-11 22:53:44 +01:00
mips pflash: Clean up after commit 368a354f02, part 2 2019-03-11 22:53:44 +01:00
misc * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
moxie moxie-softmmu.mak: express dependencies with Kconfig 2019-03-07 21:46:19 +01:00
net e1000: Delay flush queue when receive RCTL 2019-03-29 15:22:18 +08:00
nios2 nios2-softmmu.mak: express dependencies with Kconfig 2019-03-07 21:46:19 +01:00
nvram trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
openrisc or1k-softmmu.mak: express dependencies with Kconfig 2019-03-07 21:46:19 +01:00
pci * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
pci-bridge kconfig: add dependencies on CONFIG_MSI_NONBROKEN 2019-03-18 09:39:57 +01:00
pci-host * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
pcmcia kconfig: introduce kconfig files 2019-03-07 21:45:53 +01:00
ppc * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
rdma * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
riscv * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
s390x * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
scsi trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
sd trace-events: Delete unused trace points 2019-03-22 16:18:07 +00:00
sh4 hw/sh4/Kconfig: r2d machine requires the rtl8139 network card 2019-03-20 11:44:13 +01:00
smbios kconfig: introduce kconfig files 2019-03-07 21:45:53 +01:00
sparc trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
sparc64 * Kconfig improvements (msi_nonbroken, imply for default PCI devices) 2019-03-28 09:18:53 +00:00
ssi ssi: express dependencies with kconfig 2019-03-07 21:45:53 +01:00
timer trace-events: Fix attribution of trace points to source 2019-03-22 16:18:07 +00:00
tpm trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
tricore - qtest fixes 2019-03-08 16:31:34 +00:00
unicore32 unicore32-softmmu.mak: express dependencies with Kconfig 2019-03-07 21:46:19 +01:00
usb ohci: don't die on ED_LINK_LIMIT overflow 2019-03-26 12:01:45 +01:00
vfio trace-events: Fix attribution of trace points to source 2019-03-22 16:18:07 +00:00
virtio trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
watchdog trace-events: Fix attribution of trace points to source 2019-03-22 16:18:07 +00:00
xen trace-events: Shorten file names in comments 2019-03-22 16:18:07 +00:00
xenpv xen: Replace few mentions of xend by libxl 2019-01-14 13:45:40 +00:00
xtensa hw: Use PFLASH_CFI0{1,2} and TYPE_PFLASH_CFI0{1,2} 2019-03-11 22:53:44 +01:00
Kconfig kconfig: add dependencies on CONFIG_MSI_NONBROKEN 2019-03-18 09:39:57 +01:00
Makefile.objs i2c: express dependencies with Kconfig 2019-03-07 21:45:53 +01:00