- 29 Aug, 2011 40 commits
-
-
K. Y. Srinivasan authored
Linux interrupt handling code generates the eoi; don't enable auto eoi. Signed-off-by:
K. Y. Srinivasan <kys@microsoft.com> Signed-off-by:
Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@suse.de>
-
K. Y. Srinivasan authored
Get rid of vmbus_on_isr() by inlining the code. Signed-off-by:
-
K. Y. Srinivasan authored
The recent checkin that add a private pointer to hv_vmbus_device_id introduced this bug in vmbus_match; fix it. Signed-off-by:
-
K. Y. Srinivasan authored
Ensure that we correctly handle racing invocations of the channel callback when the channel is being closed. We do this using the channel's inbound_lock. A side-effect of this strategy is that we avoid repeatedly picking up this lock as we drain the inbound ring-buffer. Signed-off-by:
-
K. Y. Srinivasan authored
When we close a channel, we set the corresponding callback function to NULL. Check before invoking the channel callback. Signed-off-by:
-
K. Y. Srinivasan authored
The current code was adjusting guest time in interrupt context; do this in process context since we may have to initiate cross-processor interrupts as part of setting time. Signed-off-by:
-
K. Y. Srinivasan authored
When the host requests a "shutdown", make sure we shutdown! Signed-off-by:
-
K. Y. Srinivasan authored
No need to copy from bounce buffer in case of a failure; cleanup the code accordingly. Signed-off-by:
-
K. Y. Srinivasan authored
When we do get fatal errors from the host, offline the device since the host has already tried all possible recovery actions. Signed-off-by:
-
K. Y. Srinivasan authored
Fix a typo in a function name. Signed-off-by:
-
K. Y. Srinivasan authored
The current VHD handler on the Windows Host does not correctly handle INQUIRY and MODE_SENSE commands with some options. Fixup srb_status in these cases since the failure is not fatal. Signed-off-by:
-
K. Y. Srinivasan authored
Cleanup storvsc_drv.c after adding the contents of hyperv_storage.h. Signed-off-by:
-
K. Y. Srinivasan authored
As part of further cleanup of storvsc, add the contents of hyperv_storage.h to storvsc_drv.c and do the necessary adjustments. Signed-off-by:
-
K. Y. Srinivasan authored
Cleanup storvsc_drv.c after adding the contents of storvsc.c. Signed-off-by:
-
K. Y. Srinivasan authored
As part of further cleanup of our storage drivers, include the content of storvsc.c into storvsc_drv.c and delete storvsc.c and do the necessary adjustments. Signed-off-by:
-
K. Y. Srinivasan authored
Optimize the bounce buffer handling in the "read" case. Signed-off-by:
-
K. Y. Srinivasan authored
Optimize bounce buffer handling for the "write" case. Signed-off-by:
-
K. Y. Srinivasan authored
Now that blkvsc driver is no longer needed, remove blkvsc_drv.c Signed-off-by:
-
K. Y. Srinivasan authored
Now, enable handling of all IDE devices by extending the storvsc device id table to handle IDE guid. As part of this cleanup Kconfig and Hyper-V Makefile to not build the IDE driver (blkvsc). Signed-off-by:
-
K. Y. Srinivasan authored
Add code to handle IDE devices using the storvsc driver. The storvsc_probe() is modified so that the storvsc driver can surface all disks presented to the guest as scsi devices using generic upper level Linux scsi drivers. Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of the refcnt field in struct netvsc_device. We implement the following logic to manage the life cycle of the device: If the device is being destroyed, we do not allow any outgoing traffic. Furthermore, if the device is being destroyed, we allow incoming traffic only to drain outgoing traffic. Note that the driver may send some book keeping messages to the host not known to upper level Linux code. Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of release_inbound_net_device() by inlining the code. Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of release_outbound_net_device() by inlining the code. Signed-off-by:
-
K. Y. Srinivasan authored
Prevent outgoing traffic when netvsc dev is destroyed. Signed-off-by:
-
K. Y. Srinivasan authored
Introduce state to manage the lifecycle of net device. Signed-off-by:
-
K. Y. Srinivasan authored
Cleanup alloc_net_device(); we can directly set the refcnt. Signed-off-by:
-
K. Y. Srinivasan authored
Inline the code for free_net_device(). Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of the reference counting in struct storvsc_device. We manage the lifecycle with the following logic: If the device is marked for destruction, we dot allow any outgoing traffic on the device. Incoming traffic is allowed only to drain pending outgoing traffic. Note that while the upper level code in Linux deals with outstanding I/Os, we may have situations on Hyper-V where some book keeping messages are sent out that the upper level Linux code may not be aware of. Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of final_release_stor_device() by inlining code. Signed-off-by:
-
K. Y. Srinivasan authored
Get rid of release_stor_device() by inlining the code. Signed-off-by:
-
K. Y. Srinivasan authored
Prevent outgoing traffic when stor dev is destroyed. Signed-off-by:
-
K. Y. Srinivasan authored
Introduce state to manage the lifecycle of stor device. This would be the basis for managing the references on the stor object. Signed-off-by:
-
K. Y. Srinivasan authored
Cleanup alloc_stor_device(), we can set the ref_count directly. Signed-off-by:
-
K. Y. Srinivasan authored
In preparation for cleaning up how we manage reference counts on the stor device, clearly distinguish why we are attempting to acquire a reference. Signed-off-by:
-
K. Y. Srinivasan authored
In preparation for cleaning up how we manage reference counts on the stor device, clearly distinguish why we are attempting to acquire a reference. Signed-off-by:
-
K. Y. Srinivasan authored
On entry into storvsc_on_io_completion() we have already acquired a reference on the stor_device; there is no need to acquire an additional reference here. Signed-off-by:
-
K. Y. Srinivasan authored
Inline the code for free_stor_device() and get rid of the function. Signed-off-by:
-
Larry Finger authored
Smatch outputs the following message: drivers/staging/rtl8192e/r8192E_cmdpkt.c +412 cmpk_message_handle_rx(70) error: buffer overflow 'priv->stats.rxcmdpkt' 4 <= 7 407 RT_TRACE(COMP_CMDPKT, "---->cmpk_message_handle_rx():" 408 "unknow CMD Element\n"); 409 return 1; 410 } 411 412 priv->stats.rxcmdpkt[element_id]++; ^^^^^^^^^^ ->stats.rxcmdpkt[] only has 4 elements, but from the switch statement in the section before we can see that element_id can go up to 7 (RX_TX_RATE_HISTORY). Reported-by:Dan Carpenter <error27@gmail.com> Signed-off-by:
Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
-
Greg Kroah-Hartman authored
This resolves a conflict with: drivers/staging/brcm80211/brcmsmac/types.h Signed-off-by: -
Linus Torvalds authored
-
