qemu/hw/usb/combined-packet.c

187 lines
6.4 KiB
C
Raw Normal View History

usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
/*
* QEMU USB packet combining code (for input pipelining)
*
* Copyright(c) 2012 Red Hat, Inc.
*
* Red Hat Authors:
* Hans de Goede <hdegoede@redhat.com>
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or(at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
#include "qemu-common.h"
#include "hw/usb.h"
#include "iov.h"
#include "trace.h"
static void usb_combined_packet_add(USBCombinedPacket *combined, USBPacket *p)
{
qemu_iovec_concat(&combined->iov, &p->iov, 0, p->iov.size);
QTAILQ_INSERT_TAIL(&combined->packets, p, combined_entry);
p->combined = combined;
}
/* Note will free combined when the last packet gets removed */
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
static void usb_combined_packet_remove(USBCombinedPacket *combined,
USBPacket *p)
{
assert(p->combined == combined);
p->combined = NULL;
QTAILQ_REMOVE(&combined->packets, p, combined_entry);
if (QTAILQ_EMPTY(&combined->packets)) {
g_free(combined);
}
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
}
/* Also handles completion of non combined packets for pipelined input eps */
void usb_combined_input_packet_complete(USBDevice *dev, USBPacket *p)
{
USBCombinedPacket *combined = p->combined;
USBEndpoint *ep = p->ep;
USBPacket *next;
int status, actual_length;
bool short_not_ok, done = false;
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
if (combined == NULL) {
usb_packet_complete_one(dev, p);
goto leave;
}
assert(combined->first == p && p == QTAILQ_FIRST(&combined->packets));
status = combined->first->status;
actual_length = combined->first->actual_length;
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
short_not_ok = QTAILQ_LAST(&combined->packets, packets_head)->short_not_ok;
QTAILQ_FOREACH_SAFE(p, &combined->packets, combined_entry, next) {
if (!done) {
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
/* Distribute data over uncombined packets */
if (actual_length >= p->iov.size) {
p->actual_length = p->iov.size;
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
} else {
/* Send short or error packet to complete the transfer */
p->actual_length = actual_length;
done = true;
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
}
/* Report status on the last packet */
if (done || next == NULL) {
p->status = status;
} else {
p->status = USB_RET_SUCCESS;
}
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
p->short_not_ok = short_not_ok;
/* Note will free combined when the last packet gets removed! */
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
usb_combined_packet_remove(combined, p);
usb_packet_complete_one(dev, p);
actual_length -= p->actual_length;
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
} else {
/* Remove any leftover packets from the queue */
p->status = USB_RET_REMOVE_FROM_QUEUE;
/* Note will free combined on the last packet! */
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
dev->port->ops->complete(dev->port, p);
}
}
/* Do not use combined here, it has been freed! */
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
leave:
/* Check if there are packets in the queue waiting for our completion */
usb_ep_combine_input_packets(ep);
}
/* May only be called for combined packets! */
void usb_combined_packet_cancel(USBDevice *dev, USBPacket *p)
{
USBCombinedPacket *combined = p->combined;
assert(combined != NULL);
USBPacket *first = p->combined->first;
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
/* Note will free combined on the last packet! */
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
usb_combined_packet_remove(combined, p);
if (p == first) {
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
usb_device_cancel_packet(dev, p);
}
}
/*
* Large input transfers can get split into multiple input packets, this
* function recombines them, removing the short_not_ok checks which all but
* the last packet of such splits transfers have, thereby allowing input
* transfer pipelining (which we cannot do on short_not_ok transfers)
*/
void usb_ep_combine_input_packets(USBEndpoint *ep)
{
USBPacket *p, *u, *next, *prev = NULL, *first = NULL;
USBPort *port = ep->dev->port;
int totalsize;
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
assert(ep->pipeline);
assert(ep->pid == USB_TOKEN_IN);
QTAILQ_FOREACH_SAFE(p, &ep->queue, queue, next) {
/* Empty the queue on a halt */
if (ep->halted) {
p->status = USB_RET_REMOVE_FROM_QUEUE;
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
port->ops->complete(port, p);
continue;
}
/* Skip packets already submitted to the device */
if (p->state == USB_PACKET_ASYNC) {
prev = p;
continue;
}
usb_packet_check_state(p, USB_PACKET_QUEUED);
/*
* If the previous (combined) packet has the short_not_ok flag set
* stop, as we must not submit packets to the device after a transfer
* ending with short_not_ok packet.
*/
if (prev && prev->short_not_ok) {
break;
}
if (first) {
if (first->combined == NULL) {
USBCombinedPacket *combined = g_new0(USBCombinedPacket, 1);
combined->first = first;
QTAILQ_INIT(&combined->packets);
qemu_iovec_init(&combined->iov, 2);
usb_combined_packet_add(combined, first);
}
usb_combined_packet_add(first->combined, p);
} else {
first = p;
}
/* Is this packet the last one of a (combined) transfer? */
totalsize = (p->combined) ? p->combined->iov.size : p->iov.size;
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
if ((p->iov.size % ep->max_packet_size) != 0 || !p->short_not_ok ||
next == NULL ||
/* Work around for Linux usbfs bulk splitting + migration */
(totalsize == 16348 && p->int_req)) {
usb_device_handle_data(ep->dev, first);
assert(first->status == USB_RET_ASYNC);
usb: Add packet combining functions Currently we only do pipelining for output endpoints, since to properly support short-not-ok semantics we can only have one outstanding input packet. Since the ehci and uhci controllers have a limited per td packet size guests will split large input transfers to into multiple packets, and since we don't pipeline these, this comes with a serious performance penalty. This patch adds helper functions to (re-)combine packets which belong to 1 transfer at the guest device-driver level into 1 large transger. This can be used by (redirection) usb-devices to enable pipelining for input endpoints. This patch will combine packets together until a transfer terminating packet is encountered. A terminating packet is a packet which meets one or more of the following conditions: 1) The packet size is *not* a multiple of the endpoint max packet size 2) The packet does *not* have its short-not-ok flag set 3) The packet has its interrupt-on-complete flag set The short-not-ok flag of the combined packet is that of the terminating packet. Multiple combined packets may be submitted to the device, if the combined packets do not have their short-not-ok flag set, enabling true pipelining. If a combined packet does have its short-not-ok flag set the queue will wait with submitting further packets to the device until that packet has completed. Once enabled in the usb-redir and ehci code, this improves the speed (MB/s) of a Linux guest reading from a USB mass storage device by a factor of 1.2 - 1.5. And the main reason why I started working on this, when reading from a pl2303 USB<->serial converter, it combines the previous 4 packets submitted per device-driver level read into 1 big read, reducing the number of packets / sec by a factor 4, and it allows to have multiple reads outstanding. This allows for much better latency tolerance without the pl2303's internal buffer overflowing (which was happening at 115200 bps, without serial flow control). Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2012-10-31 16:47:09 +04:00
if (first->combined) {
QTAILQ_FOREACH(u, &first->combined->packets, combined_entry) {
usb_packet_set_state(u, USB_PACKET_ASYNC);
}
} else {
usb_packet_set_state(first, USB_PACKET_ASYNC);
}
first = NULL;
prev = p;
}
}
}