Commit graph

321,296 commits

Author SHA1 Message Date
Tony Zelenoff
2a9bc71e9a atl1: do not process interrupts in cycle in handler
As the rx/tx handled inside napi handler, the cycle is
not needed now, because only the rx/tx need such kind of
processing.

Signed-off-by: Tony Zelenoff <antonz@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 13:00:12 -04:00
Tony Zelenoff
73650f28ae atl1: enable errors and link ints when rx/tx scheduled
Signed-off-by: Tony Zelenoff <antonz@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:56:02 -04:00
Tony Zelenoff
aa45ba90b5 atl1: add value to check ability of reenabling IRQs
Unfortunately it is not clear from code is usage of
IMR register possible or not. So, to prevent possible
side-effects of reading this register i prefer store
interrupts enable flag separately.

Signed-off-by: Tony Zelenoff <antonz@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:56:02 -04:00
Tony Zelenoff
02d5d11bfa atl1: make function to set imr of card
This function should be used later to set/remove proper
bits in imr to disable only rx ints.

Signed-off-by: Tony Zelenoff <antonz@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:56:02 -04:00
Tony Zelenoff
5c3d52ef5a atl1: use defined functions to disable irq
Looks like direct writes to IMR register is not good idea,
because there are exist functions to make this work.

Signed-off-by: Tony Zelenoff <antonz@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:56:01 -04:00
Tony Zelenoff
0dbab2fb1d atl1: add napi process of tx interrupts
Make the tx ints processing same as rx ones via napi.
The idea got from e1000. The interrupt disabling is
still not fine grained.

Signed-off-by: Tony Zelenoff <antonz@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:55:08 -04:00
Tony Zelenoff
6294512bbe atl1: make driver napi compatible
This is first step, here there is no fine interrupt
disabling which cause TX/ERR interrupts stalling when
RX scheduled ints processed.

Signed-off-by: Tony Zelenoff <antonz@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:55:07 -04:00
Tony Zelenoff
3e1d83f711 atl1: handle rx in separate condition
Remove rx from unlikely optimization in case of rx is very
likely thing for network card. This also reduce code a bit.

Signed-off-by: Tony Zelenoff <antonz@parallels.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:54:15 -04:00
Herbert Xu
c5c2326059 bridge: Add multicast_querier toggle and disable queries by default
Sending general queries was implemented as an optimisation to speed
up convergence on start-up.  In order to prevent interference with
multicast routers a zero source address has to be used.

Unfortunately these packets appear to cause some multicast-aware
switches to misbehave, e.g., by disrupting multicast packets to us.

Since the multicast snooping feature still functions without sending
our own queries, this patch will change the default to not send
queries.

For those that need queries in order to speed up convergence on start-up,
a toggle is provided to restore the previous behaviour.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:51:35 -04:00
Herbert Xu
c83b8fab06 bridge: Restart queries when last querier expires
As it stands when we discover that a real querier (one that queries
with a non-zero source address) we stop querying.  However, even
after said querier has fallen off the edge of the earth, we will
never restart querying (unless the bridge itself is restarted).

This patch fixes this by kicking our own querier into gear when
the timer for other queriers expire.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:51:34 -04:00
Herbert Xu
748572162a bridge: Add br_multicast_start_querier
This patch adds the helper br_multicast_start_querier so that
the code which starts the queriers in br_multicast_toggle can
be reused elsewhere.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:51:34 -04:00
Eric Dumazet
95c9617472 net: cleanup unsigned to unsigned int
Use of "unsigned int" is preferred to bare "unsigned" in net tree.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:44:40 -04:00
Daniel Baluta
5e73ea1a31 ipv4: fix checkpatch errors
Fix checkpatch errors of the following type:
	* ERROR: "foo * bar" should be "foo *bar"
	* ERROR: "(foo*)" should be "(foo *)"

Signed-off-by: Daniel Baluta <dbaluta@ixiacom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 12:37:19 -04:00
Wang YanQing
35fca53e15 mmzone: fix comment typo coelesce -> coalesce
Signed-off-by: Wang YanQing <udknight@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2012-04-15 16:57:23 +02:00
Mark Hills
7536c301f8 ALSA: snd-usb-audio: Replace mixer for Electrix Ebox-44
The mixer units from the firmware are corrupt, and even where they
are valid they presents mono controls as L and R channels of
stereo.

Signed-off-by: Mark Hills <mark@pogo.org.uk>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2012-04-15 15:40:08 +02:00
Mark Hills
284a8dd6f0 ALSA: snd-usb-audio: Skip un-parseable mixer units instead of erroring
Some interfaces reference endpoints which do not exists. To
accomodate these, do not fail completely, but skip over them.

This allows the Electrix Ebox-44 with earlier firmware to be
detected and used for audio.

Signed-off-by: Mark Hills <mark@pogo.org.uk>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2012-04-15 15:39:55 +02:00
Mark Brown
9f29c9e30b regulator: wm8994: Convert to set_voltage_sel()
There's no need to implement set_voltage() as there is only a very small
range of selectors for the regulators in the WM8994 and the voltages are
not expected to vary frequently at runtime.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
2012-04-15 12:07:50 +01:00
Dan Carpenter
60884c2767 ASoC: dapm: release lock on error paths
We added locking here but there were a couple error paths where we
forgot to drop the lock before returning.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
2012-04-15 10:46:17 +01:00
Michael S. Tsirkin
3ccc9372ed virtio_balloon: fix handling of PAGE_SIZE != 4k
As reported by David Gibson, current code handles PAGE_SIZE != 4k
completely wrong which can lead to guest memory corruption errors:

- page_to_balloon_pfn is wrong: e.g. on system with 64K page size
 it gives the same pfn value for 16 different pages.

- we also need to convert back to linux pfns when we free.

- for each linux page we need to tell host about multiple balloon
  pages, but code only adds one pfn to the array.

This patch fixes all that, tested with a 64k ppc64 kernel.

Reported-by: David Gibson <david@gibson.dropbear.id.au>
Tested-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2012-04-15 11:51:06 +03:00
David Gibson
1a87228f5f virtio_balloon: Fix endian bug
Although virtio config space fields are usually in guest-native endian,
the spec for the virtio balloon device explicitly states that both fields
in its config space are little-endian.

However, the current virtio_balloon driver does not have a suitable endian
swap for the 'num_pages' field, although it does have one for the 'actual'
field.  This patch corrects the bug, adding sparse annotation while we're
at it.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2012-04-15 11:51:05 +03:00
Jason Wang
586d17c5a0 virtio-net: send gratuitous packets when needed
As hypervior does not have the knowledge of guest network configuration, it's
better to ask guest to send gratuitous packets when needed.

This patch implements VIRTIO_NET_F_GUEST_ANNOUNCE feature: hypervisor would
notice the guest when it thinks it's time for guest to announce the link
presnece. Guest tests VIRTIO_NET_S_ANNOUNCE bit during config change interrupt
and woule send gratuitous packets through netif_notify_peers() and ack the
notification through ctrl vq.

We need to make sure the atomicy of read and ack in guest otherwise we may ack
more times than being notified. This is done through handling the whole config
change interrupt in an non-reentrant workqueue.

Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 03:23:31 -04:00
Peter Hüwe
8831a3f2c9 isdn/hysdn: Convert to kstrtoul_from_user
This patch replaces the code for getting an number from a
userspace buffer by a simple call to kstroul_from_user.
This makes it easier to read and less error prone.

Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-15 03:23:31 -04:00
Ingo Molnar
7ea6411f4c Fixes for perf/urgent:
. Properly handle ~/.debug, the build id cache, when it is a symlink,
   fix from Chanho Park
 
 . Fixes for the parser generation process, from Jiri Olsa and Namhyung Kim
 
 . Fix build when NO_GTK2 is specified, From Stephane Eranian
 
 . When a machine is not found, bump the relevant error stat but return
   0, so that we correctly move to the next perf event. Fix from Jiri Olsa
 
 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2.0.14 (GNU/Linux)
 
 iQIcBAABAgAGBQJPiayHAAoJENZQFvNTUqpAU+MP/RSDuLTNxhYf0VDvKi55gKnQ
 NK+1IjfnBbEo3uz2/WnlaXfj0TtiCNp0+r0dQqn3Hzuw9sc7zVOE+V1KlPvKMSml
 ruFn1NHEAc4J+/pa5t/tNEzFUm4IhtemclroWA+NplVS9trVzfVwwypieVOe+wkM
 ICRke30dfY+xvHOk+dAo3PQ/5qiycRH2GF4PYgivdU0psGPrTIbXICEjf0LRZzIZ
 B0a41Q9gASKNEXWQ6jEH7eGSyAZHuJ3V/tt4y1Nlnh94AgUwogHMohKf+ASTFI52
 BorwuDECIJGi1X9pdaP0A/Y6IdCNK0/sNGWl06FKW/wQ4qCpBucZrVv+CRkQX3B5
 vq23CrmOSNy6KH6s63z05v/v/v8mdnhofCRITXwztjnsdPDTosj742b/BKc7y8HY
 RY5mxinJgJptWRoBvpkVT/+xqn2Pn9Wup2XyM7vQAMoqqV7otpuGMQZ6V/ozymR2
 aBWzT8O+JerYmTb3HoBqjHSUIgTENGTawCYoXY3YPPH6gz9IDfPqCiyKRyWIhPnP
 Iw9L5gWQIodJNsmRfuK9qt4sTbMl7XvfXuyFYb2/YANNDkR7OyQktJnXMF3jj1cJ
 yAXV7sMPv4OWCDUBAgtDLsJHOuSovImM7M10UpzGnBKRNnYrjf6VT9BdWMQD22lP
 QDn7GRoElJT2tGn5CfeV
 =eQKW
 -----END PGP SIGNATURE-----

Merge tag 'perf-urgent-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent

Pull perf tooling fixes from Arnaldo Carvalho de Melo:

. Properly handle ~/.debug, the build id cache, when it is a symlink,
  fix from Chanho Park

. Fixes for the parser generation process, from Jiri Olsa and Namhyung Kim

. Fix build when NO_GTK2 is specified, From Stephane Eranian

. When a machine is not found, bump the relevant error stat but return
  0, so that we correctly move to the next perf event. Fix from Jiri Olsa

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-04-15 08:02:36 +02:00
Olof Johansson
17d2f847e6 Merge branch 'ixp2xxx/removal' into next/cleanup
* ixp2xxx/removal:
  ARM: remove ixp23xx and ixp2000 platforms
2012-04-14 22:17:47 -07:00
David S. Miller
cf22f9a2b8 ipv6: Remove unused argument to addrconf_dad_start().
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-14 21:37:40 -04:00
Olof Johansson
961a238c38 Merge branch 'v3.4-samsung-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung into fixes
* 'v3.4-samsung-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/kgene/linux-samsung:
  ARM: EXYNOS: Fix Kconfig dependencies for device tree enabled machine files
  ARM: EXYNOS: Remove broken config values for touchscren for NURI board
  ARM: EXYNOS: set fix xusbxti clock for NURI and Universal210 boards
  ARM: EXYNOS: fix regulator name for NURI board
  ARM: SAMSUNG: make SAMSUNG_PM_DEBUG select DEBUG_LL
  ARM: EXYNOS: Add PDMA and MDMA physical base address defines
  ARM: S5PV210: Fix compiler warning in dma.c file
  ARM: EXYNOS: Fix compile error in exynos5250-cpufreq.c
  ARM: EXYNOS: Add missing definition for IRQ_I2S0
  ARM: S5PV210: fix unused LDO supply field from wm8994_pdata
2012-04-14 17:50:45 -07:00
Kevin Hilman
47d9e44d1d ARM: OMAP: clock: cleanup CPUfreq leftovers, fix build errors
Now that we have OPP layer, and OMAP CPUfreq driver is using it, we no
longer need/use the clock framework code for filling up CPUfreq
tables.  Remove it.

Removing this code also eliminates build errors when CPU_FREQ_TABLE
support is not enabled.

Thanks to Russell King for pointing out the parts I missed under
plat-omap in the original version and also pointing out the build
errors when CPUFREQ_TABLE support was not enabled.

Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Kevin Hilman <khilman@ti.com>
Acked-by: Paul Walmsley <paul@pwsan.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Olof Johansson <olof@lixom.net>
2012-04-14 17:49:07 -07:00
Rob Herring
eb03f28002 ARM: dts: remove blank interrupt-parent properties
These were incorrectly introduced and can cause problems for of_irq_init.

The correct way to define a root controller is no interrupt-parent set at
all or the interrupt-parent is set to the root controller itself when
inherited from a parent node.

Signed-off-by: Rob Herring <rob.herring@calxeda.com>
Tested-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Olof Johansson <olof@lixom.net>
2012-04-14 17:46:46 -07:00
Olof Johansson
9e7bcbe8be Merge branch 'msm-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/davidb/linux-msm into fixes
From David Brown:
  "Here are some fixes for msm that fix problems caused by the latest
   ARM code.  The ones from Daniel remove unneeded fixups that now
   cause compilation failures.  Mine fix section mismatches, that were
   incompletely fixed earlier."

* 'msm-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/davidb/linux-msm:
  ARM: msm: Fix section mismatches in proc_comm.c
  video: msm: Fix section mismatches in mddi.c
  arm: msm: trout: fix compile failure
  arm: msm: halibut: remove unneeded fixup
2012-04-14 17:44:21 -07:00
Andy Grover
bfb79eac20 target/iscsi: Go back to core allocating data buffer for cmd
We originally changed iscsi to allocate its own buffers just as an
intermediate step to clean up some core buffer allocation mechanisms. Now
we can put it back.

Also had to change allocate_iovecs to use data_length instead of
t_data_nents because iovecs are now allocated before the data buffer, thus
t_data_nents is not yet initialized.

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:39 -07:00
Andy Grover
11e319ed95 target: Call core_alua_check_nonop_delay in target_submit_cmd()
It appears iscsi is the only one to call this in its cmd submit path, but
it appears to be applicable to all fabrics, and should always be called.

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:39 -07:00
Andy Grover
4334e49bca target/iscsi: Fold _decide_list_to_build into _build_pdu_and_seq_lists
Rename iscsit_build_pdu_and_seq_list to iscsit_do_build_pdu_and_seq_lists

Rename iscsit_do_build_list to iscsit_build_pdu_and_seq_lists

Move code from iscsit_decide_list_to_build into _seq_pdu_list.c, seems
a better fit.

Also update some comments in pdu/seq code for correctness and whitespace.

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:38 -07:00
Andy Grover
ebf1d95ca2 target/iscsi: Eliminate iscsi_cmd.data_length
Redundant, just use iscsi_cmd->se_cmd.data_length once se_cmd is
initialized, or hdr->data_length before then.

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:38 -07:00
Andy Grover
065ca1e42f target/iscsi: Move init_se_cmd closer to lookup_cmd_lun
if we can get calls to init_se_cmd, get_sess_cmd, lookup_cmd_lun,
core_alua_check_nonop_delay, and handle_cdb_direct next to each other,
then we can just call target_submit_cmd. This is a step towards that
goal.

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:37 -07:00
Andy Grover
d28b11692e target/iscsi: Inline iscsit_allocate_se_cmd and *_for_tmr
Trying to move a bunch of stuff around so iscsi can use target_submit_cmd
someday, and so stuff needs to be in that function directly instead of
hidden, so it can be reordered etc.

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:37 -07:00
Andy Grover
b16a35b050 target: rewrite comment for generic_new_cmd
Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:36 -07:00
Andy Grover
a12f41f841 target: Rename target_allocate_tasks to target_setup_cmd_from_cdb
This patch renames a horribly misnamed function that no longer allocate
tasks to something more descriptive for it's modern use in target core.

(nab: Fix up ib_srpt to use this as well ahead of a target_submit_cmd
conversion)

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:36 -07:00
Andy Grover
8b1e1244db target/iscsi: Misc cleanups from Agrover (round 2)
This patch includes the handful of squashed patches for target/iscsi from
Andy's original series into lio-core/master code:

*) Make iscsit_add_reject static
*) Remove unused data_offset_end from iscsi_datain_req
*) Remove "#if 0" stubs
*) Rename iscsi_datain_req to cmd_datain_node
*) Cleanups for built_r2ts_for_cmd()
*) Cleanups for Cleanup build_sendtargets_response()

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:35 -07:00
Andy Grover
6f3c0e69a9 target/iscsi: Refactor target_tx_thread immediate+response queue loops
Immediate queue:

Consolidate down to one switch statement by moving send_tx_data and stuff
from second switch into the first switch, or the functions the first switch
calls.

Response queue:

Do not lock istate_lock except directly around i_state modifications.

Put entire ISTATE_SEND_DATAIN path within first switch statement, in prep
for further refactoring.

All other cases set use_misc = 1 and will not be using sendpage, so just
use send_tx_data for these and set use_misc param to 1.

map_sg, sent_status, use_misc, and se_cmd vars no longer needed.

Then put immediate and response handling in separate functions in order
to get iscsi_target_tx_thread down to where it fits on a page.

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:34 -07:00
Andy Grover
4580cf3848 target/iscsi: Remove unneeded locking from iscsi_target_tx_thread
When processing immediate queue, we're switching on a local variable
so it's not necessary to lock around it. However, we are modifying
cmd->i_state in two spots, so lock around those parts only.

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:34 -07:00
Andy Grover
c6037cc546 target/iscsi: Misc cleanups from Agrover (round 1)
*) Use decoded cmd->immediate_cmd for conditional instead of
   re-examining hdr->opcode
*) Make iscist_dataout_post_crc_passed more legible
*) use max() to reduce code in build_r2ts_for_cmd()
*) Remove CONFIG_SMP and if 0 ifdefs
*) Replace if/goto with a while loop
*) Remove unused conn->tx_immediate_queue and tx_response_queue

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:33 -07:00
Andy Grover
2fbb471e78 target/iscsi: Rename iscsi_cmd.i_list to iscsi_cmd.i_conn_node
The name change makes it clear this list_head is so the cmd can be an
item in the connection's conn_cmd_list.

Signed-off-by: Andy Grover <agrover@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:33 -07:00
Roland Dreier
957525a242 target: Remove transport_do_task_sg_chain() and associated detritus
Now that all fabrics are converted over to using se_cmd->t_data_sg
directly, we can drop the task sg chaining support.  With the modern
memory allocation in target core, task sg chaining is needless
overhead -- we would split up the main cmd sglist into pieces, and
then splice those pieces back together instead of just using the
original list directly.

Signed-off-by: Roland Dreier <roland@purestorage.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:32 -07:00
Roland Dreier
6f9e7f01b6 IB/srpt: Remove use of transport_do_task_sg_chain()
With the modern target core, se_cmd->t_data_sg already points to a
sglist that covers the whole command.  So task_sg chaining is needless
overhead and obfuscation -- instead of splicing the split up task
sglists back into one list, we can just use the original list directly.

Signed-off-by: Roland Dreier <roland@purestorage.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:32 -07:00
Roland Dreier
e182d6828d tcm_fc: Remove use of transport_do_task_sg_chain()
With the modern target core, se_cmd->t_data_sg already points to a
sglist that covers the whole command.  So task_sg chaining is needless
overhead and obfuscation -- instead of splicing the split up task
sglists back into one list, we can just use the original list directly.

Signed-off-by: Roland Dreier <roland@purestorage.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Kiran Patil <Kiran.patil@intel.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:31 -07:00
Christoph Hellwig
bebe2fdc17 target: stop splitting commands into multiple tasks
From hch:

The high-performance backends (iblock and rd) support tasks of unlimited
size.  With that there is no reason to keep a complex infrastructure for
splitting up commands in place.  Stop doing so and only submit a single
task per data direction.  Once this is in place we can slowly remove fields
from the task that duplicate things in the command, or move other fields
into the command.

From nab:

The benefit to IBLOCK performance by removing the additional
fast-path allocation overhead + SGL mapping to se_task->task_sg[] is now
greater than transparently supporting an received CDB I/O length that
exceeds what is allowed by backend pSCSI LLD hardware max_sectors, that
was originally supported for all backend export cases.

This change may effect some users of pSCSI users on legacy hardware, but
I think most folks are now using TYPE_DISK struct scsi_device export
with IBLOCK.  The only other place where this may can issues that cannot
be resolved with IBLOCK TYPE_DISK is using TYPE_ROM, TYPE_TAPE or other
pSCSI non TYPE_DISK export with an SCSI LLDs using a smaller
max_sectors.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:31 -07:00
Christoph Hellwig
83799efbff target: don't limit transfer sizes for the ramdisk backend
The ramdisk backend has not inherent limitations for handling requests,
so don't artificially limits the transfer size.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:30 -07:00
Christoph Hellwig
8feb58d04b target: misc ramdisk backend cleanups
Remove various leftovers of the old direct/indirect split, as well as the
unused rd_request structure and a couple unused defines and fields.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:30 -07:00
Sebastian Andrzej Siewior
b0d7994660 target: add unknown size flag to target_submit_cmd()
The UASP protocol does not inform the target device upfront how much
data it should expect so we have to learn in from the CDB.  So in order
to handle this case, add a TARGET_SCF_UNKNOWN_SIZE to target_submit_cmd()
and perform an explictly assignment for se_cmd->data_length from the
extracted CDB size in transport_generic_cmd_sequencer().

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 17:40:03 -07:00
Nicholas Bellinger
58d9261877 target: Remove obsolete DF_READ_ONLY usage
This was used at one time as a hack by FILEIO backend registration to
allow a struct block_device that was claimed with blkdev_get (by a local
filesystem mount for example) to be exported as read-only (SCSI WP=1).

Since FILEIO backend registration will no longer attempt to obtain
exclusive access to an underlying struct block_device here, this flag is
now obsolete.

Reported-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2012-04-14 16:04:58 -07:00