Commit graph

388,156 commits

Author SHA1 Message Date
Nithin Sujir
837c45bb4e tg3: Refactor cpu pause/resume code
The 57766 rxcpu needs to be paused/resumed when we download the firmware just
like we do for existing firmware. Refactor the pause/resume code to be
reusable.

This patch also renames the "offset" argument of tg3_halt_cpu to "cpu_base"
since that's what it really is.

Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: Nithin Nayak Sujir <nsujir@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-07 15:31:31 -05:00
Matt Carlson
1caf13ebc8 tg3: Add new FW_TSO flag
tg3 used the fw_needed member loosely as a synonym for firmware TSO. Now
that the 57766 needs firmware download support, fw_needed can no longer be
used like this. This patch creates a new FW_TSO flag and changes the
code to use it.

Also rearrange all the TSO flags together in the enum.

Reviewed-by: Benjamin Li <benli@broadcom.com>
Signed-off-by: Nithin Nayak Sujir <nsujir@broadcom.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-07 15:31:31 -05:00
David S. Miller
a6f3587ca2 Merge branch 'fdb'
Vlad Yasevich says:

====================
This is a short series that now allows mac filter programming on any
card that support IFF_UNICAST_FLT by using the existing FDB interface.

Some existing drivers that had FDB functionality usually supported
it only in SR-IOV mode.  Since that's not always enabled, and
we want to take advantage of IFF_UNICAST_FLT support, these drivers
have been converted to call the default handler when not in SRIOV mode.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-07 15:29:54 -05:00
Vlad Yasevich
3e5c112f5f qlcnic: Use generic fdb handler when driver options are not enabled.
Allow qlcnic to use the generic fdb handler when the driver options
are not enabled.   Untill the driver is fully fixed, this allows
the use of the FDB interface with qlogic driver, but simply puts
the driver into promisc mode since the driver currently does not
support IFF_UNICAST_FLT.

CC: Jitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Acked-by: Jitendra Kalsaria <jitendra.kalsaria@qlogic.com>
CC: Sony Chacko <sony.chacko@qlogic.com>
CC: linux-driver@qlogic.com
Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-07 15:29:46 -05:00
Vlad Yasevich
75a75ee46b mlx4: Remove driver specific fdb handlers.
Remove driver specific fdb hadlers since they are the same as
the default ones.

CC: Amir Vadai <amirv@mellanox.com>
CC: Yan Burman <yanb@mellanox.com>
Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-07 15:29:46 -05:00
Vlad Yasevich
faaf02d24c ixgbe: Make use of the default fdb handlers.
For fdb_add, use the default handler in the non-SRIOV case.
For the other fdb handlers, just remove them and use the
default ones.

CC: John Fastabend <john.r.fastabend@intel.com>
Acked-By: John Fastabend <john.r.fastabend@intel.com>
CC: CC: Gregory Rose <gregory.v.rose@intel.com>
Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-07 15:29:45 -05:00
Vlad Yasevich
090096bf3d net: generic fdb support for drivers without ndo_fdb_<op>
If the driver does not support the ndo_op use the generic
handler for it. This should work in the majority of cases.
Eventually the fdb_dflt_add call gets translated into a
__dev_set_rx_mode() call which should handle hardware
support for filtering via the IFF_UNICAST_FLT flag.

Namely IFF_UNICAST_FLT indicates if the hardware can do
unicast address filtering. If no support is available
the device is put into promisc mode.

Signed-off-by: Vlad Yasevich <vyasevic@redhat.com>
Signed-off-by: John Fastabend <john.r.fastabend@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-07 15:29:45 -05:00
Takuya Yoshikawa
5da596078f KVM: MMU: Introduce a helper function for FIFO zapping
Make the code for zapping the oldest mmu page, placed at the tail of the
active list, a separate function.

Reviewed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 17:26:27 -03:00
Takuya Yoshikawa
945315b9db KVM: MMU: Use list_for_each_entry_safe in kvm_mmu_commit_zap_page()
We are traversing the linked list, invalid_list, deleting each entry by
kvm_mmu_free_page().  _safe version is there for such a case.

Reviewed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 17:26:27 -03:00
Takuya Yoshikawa
1044b03034 KVM: MMU: Fix and clean up for_each_gfn_* macros
The expression (sp)->gfn should not be expanded using @gfn.
Although no user of these macros passes a string other than gfn now,
this should be fixed before anyone sees strange errors.

Note: ignored the following checkpatch errors:
  ERROR: Macros with complex values should be enclosed in parenthesis
  ERROR: trailing statements should be on next line

Reviewed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 17:26:27 -03:00
Daniel Pieczko
1648a23fa1 sfc: allocate more RX buffers per page
Allocating 2 buffers per page is insanely inefficient when MTU is 1500
and PAGE_SIZE is 64K (as it usually is on POWER).  Allocate as many as
we can fit, and choose the refill batch size at run-time so that we
still always use a whole page at once.

[bwh: Fix loop condition to allow for compound pages; rebase]
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:15 +00:00
Ben Hutchings
179ea7f039 sfc: Replace efx_rx_is_last_buffer() with a flag
This condition is brittle and we have lots of flags to spare.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:14 +00:00
Daniel Pieczko
2768935a46 sfc: reuse pages to avoid DMA mapping/unmapping costs
On POWER systems, DMA mapping/unmapping operations are very expensive.
These changes reduce these costs by trying to reuse DMA mapped pages.

After all the buffers associated with a page have been processed and
passed up, the page is placed into a ring (if there is room).  For
each page that is required for a refill operation, a page in the ring
is examined to determine if its page count has fallen to 1, ie. the
kernel has released its reference to these packets.  If this is the
case, the page can be immediately added back into the RX descriptor
ring, without having to re-map it for DMA.

If the kernel is still holding a reference to this page, it is removed
from the ring and unmapped for DMA.  Then a new page, which can
immediately be used by RX buffers in the descriptor ring, is allocated
and DMA mapped.

The time a page needs to spend in the recycle ring before the kernel
has released its page references is based on the number of buffers
that use this page.  As large pages can hold more RX buffers, the RX
recycle ring can be shorter.  This reduces memory usage on POWER
systems, while maintaining the performance gain achieved by recycling
pages, following the driver change to pack more than two RX buffers
into large pages.

When an IOMMU is not present, the recycle ring can be small to reduce
memory usage, since DMA mapping operations are inexpensive.

With a small recycle ring, attempting to refill the descriptor queue
with more buffers than the equivalent size of the recycle ring could
ultimately lead to memory leaks if page entries in the recycle ring
were overwritten.  To prevent this, the check to see if the recycle
ring is full is changed to check if the next entry to be written is
NULL.

[bwh: Combine and rebase several commits so this is complete
 before the following buffer-packing changes.  Remove module
 parameter.]
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:13 +00:00
Ben Hutchings
85740cdf0b sfc: Enable RX DMA scattering where possible
Enable RX DMA scattering iff an RX buffer large enough for the current
MTU will not fit into a single page and the NIC supports DMA
scattering for kernel-mode RX queues.

On Falcon and Siena, the RX_USR_BUF_SIZE field is used as the DMA
limit for both all RX queues with scatter enabled.  Set it to 1824,
matching what Onload uses now.

Maintain a statistic for frames truncated due to lack of descriptors
(rx_nodesc_trunc).  This is distinct from rx_frm_trunc which may be
incremented when scattering is disabled and implies an over-length
frame.

Whenever an MTU change causes scattering to be turned on or off,
update filters that point to the PF queues, but leave others
unchanged, as VF drivers assume scattering is off.

Add n_frags parameters to various functions, and make them iterate:
- efx_rx_packet()
- efx_recycle_rx_buffers()
- efx_rx_mk_skb()
- efx_rx_deliver()

Make efx_handle_rx_event() responsible for updating
efx_rx_queue::removed_count.

Change the RX pipeline state to a starting ring index and number of
fragments, and make __efx_rx_packet() responsible for clearing it.

Based on earlier versions by David Riddoch and Jon Cooper.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:12 +00:00
Ben Hutchings
b74e3e8cd6 sfc: Update RX buffer address together with length
Adjust rx_buf->page_offset when we eat the RX hash prefix.  Remove
efx_rx_buf_offset(), which is now redundant.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:11 +00:00
Ben Hutchings
5036b7c7b9 sfc: Explicitly prefetch RX hash prefix, not just Ethernet heade
Currently we prefetch from the Ethernet header, but we will also read
the hash prefix.  In practice they should be in the same cache line
and this won't hurt, but it is still pointless to add on the hash
prefix size.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:10 +00:00
Ben Hutchings
b184f16b7f sfc: Replace efx_rx_buf_eh() with simpler efx_rx_buf_va()
efx_rx_buf_va() returns the virtual address of the current start of
the buffer.  The callers must add the hash prefix size themselves.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:09 +00:00
Ben Hutchings
ff734ef4bc sfc: Wrap __efx_rx_packet() with efx_rx_flush_packet()
The pipeline mechanism will need to change a bit for scattered
packets.  Add a wrapper to insulate efx_process_channel() from this.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:08 +00:00
Ben Hutchings
9bc2fc9b52 sfc: Make RX queue descriptor counts unsigned for consistency
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:07 +00:00
Ben Hutchings
272baeeb6a sfc: Properly distinguish RX buffer and DMA lengths
Replace efx_nic::rx_buffer_len with efx_nic::rx_dma_len, the maximum
RX DMA length.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:06 +00:00
Ben Hutchings
80c2e716d5 sfc: Document current usage of efx_rx_buffer::len and efx_nic::rx_buffer_len
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:05 +00:00
Alexandre Rames
626950db84 sfc: Add AER and EEH support for Siena
The Linux side of EEH is triggered by MMIO reads, but this
driver's data path does not issue any MMIO reads (except in
legacy interrupt mode).  Therefore add a monitor function
to poll EEH periodically.

When preparing to reset the device based on our own error
detection, also poll EEH and defer to its recovery mechanism
if appropriate.

[bwh: Use a separate condition for the initial link poll; fix some
 style errors]
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:04 +00:00
Ben Hutchings
634ab72c39 sfc: Disable RSS when using SR-IOV and only 1 RX queue on the PF
On Siena, VFs share RSS configuration with the PF.  We attempted to
support configurations where the PF only uses 1 RX queue and VFs use
multiple RX queues, by (1) setting up RSS for the number of RX queues
per VF (2) disabling RSS in the PF's RX default filters.

Unfortunately commit cd2d5b529c ('sfc: Add SR-IOV back-end support
for SFC9000 family') only included (1).  This is (2).

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:03 +00:00
Ben Hutchings
d9ccfdd4b3 sfc: Fix replacement detection in efx_filter_insert_filter()
efx_filter_insert_filter() uses the first table entry in the hash chain
that either has the same match values or is empty.  This means that
replacement doesn't always work correctly:

1. Insert filter F1 with match values M1, hashing to H1, at first
   possible entry E1.
2. Insert filter F2 with match values M2, hashing to H1, at second
   possible entry E2.
3. Remove filter F1.
4. Insert filter F3 with match values M2, hashing to H1, at first
   possible entry E1.

F3 should have either replaced F2 or been rejected (depending on
priority and the replace_equal parameter).

Instead, search for both a matching filter that the inserted filter
would replace, and an available insertion point, up to the applicable
maximum search depths.  If we insert at lower depth than a replaced
filter, clear the replaced filter.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:02 +00:00
Ben Hutchings
297891cecc sfc: Merge efx_filter_search() into efx_filter_insert()
efx_filter_search() is only called from efx_filter_insert(), and
neither function is very long.  The following bug fix requires a more
sophisticated search with a third result, which is going to be easier
to implement as part of the same function.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:01 +00:00
Ben Hutchings
385904f819 sfc: Don't use efx_filter_{build,hash,increment}() for default MAC filters
These functions happen to work for default MAC filters: they generate
an initial index of 1/0 for unicast/multicast respectively and an
increment of 1 for either, so a search succeeds at depth 2.  But this
is a matter of luck rather than design, and it really won't work well
with the bug fix we're about to do.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:22:00 +00:00
Ben Hutchings
e3a699fab3 sfc: Remove redundant parameter to efx_filter_search()
The 'for_insert' parameter is redundant since there are no longer
any other operations that need to search based on a filter spec.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:21:59 +00:00
Ben Hutchings
7de07a4deb sfc: More sensible semantics for efx_filter_insert_filter() replace flag
The 'replace' flag to efx_filter_insert_filter() controls whether the
new filter may replace *any* filter, and is checked even before
priority comparison.  But lower-priority filters should never
block insertion of higher-priority filters.

Change the priority checking so that lower-priority filters are
replaced regardless of the value of the flag, and rename the
flag to 'replace_equal'.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:21:58 +00:00
Alexandre Rames
97d48a10c6 sfc: Remove rx_alloc_method SKB
[bwh: Remove more dead code, and make efx_ptp_rx() pull the data it
 needs into the header area.]
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:21:57 +00:00
Laurence Evans
9230451af9 sfc: tidy up PTP synchronize function efx_ptp_process_times()
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:21:56 +00:00
Laurence Evans
c939a31645 sfc: PTP changes to support improved UUID filtering mode
There is a long-standing problem with the packet-timestamp matching in
the driver. When a PTP packet is received by the MC, the FPGA
timestamps the packet and the MC sends the timestamp and 6 bytes of
the UUID to the driver. The driver then matches the timestamp against
received packets using the same 6 bytes of UUID.

The problem comes from the choice of which 6 bytes to use. The PTP
spec is slightly contradictory and misleading in one of the two places
where the UUIDs are discussed. From section 7.2.2.2 of the spec, a
PTPD2 UUID can be either a EUI-64 or a EUI-64 constructed from a
EUI-48. The typical ethernet based implementation uses a EUI-64
constructed from a EUI-48. This works by taking the first 3 bytes of
the MAC address of the NIC being used for PTP (the OUI), then
inserting 0xFF, 0xFE, then taking the last 3 bytes of the MAC address
giving
          MAC[0], MAC[1], MAC[2], 0xFF, 0xFE, MAC[3], MAC[4], MAC[5]
The current MC firmware and driver discard the first two bytes of this
UUID and packets are matched against timestamps using bytes 2 to 7 so
there is a small risk that in a deployment of Solarflare PTP NICs used
with other vendors NICs, that a PTP packet could be matched against
the wrong timestamp. This applies to all other organisations whose
third byte of the OUI is 0x53. It's a long list but I notice that it
includes Cisco.

The necessary modifications to use bytes 0-2 and 5-7 of the UUID to
match against are quite small but introduce incompatibility between
older version of the firmware and driver.

When PTP is enabled via SO_TIMESTAMPING specifying PTP V2, the driver
will try to enable PTP in the firmware using the enhanced mode
(above). If the firmware returns an error, the driver will enable PTP
in the firmware using the old mode.

[bwh: Fix some style errors; remove private ioctl bits]
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:21:55 +00:00
Ben Hutchings
4a74dc65e3 sfc: Allow efx_channel_type::receive_skb() to reject a packet
Instead of having efx_ptp_rx() call netif_receive_skb() for an invalid
PTP packet, make it return false for rejected packets and have
efx_rx_deliver() pass them up.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
2013-03-07 20:21:54 +00:00
Ben Hutchings
86c2da58a7 Merge branch 'sfc-3.9' into master 2013-03-07 20:21:30 +00:00
David S. Miller
43b18db8a2 Merge branch 'master' of git://1984.lsi.us.es/nf
Pablo Neira Ayuso says:

====================
The following patchset contains Netfilter fixes for your net tree,
they are:

* Don't generate audit log message if audit is not enabled, from Gao Feng.

* Fix logging formatting for packets dropped by helpers, by Joe Perches.

* Fix a compilation warning in nfnetlink if CONFIG_PROVE_RCU is not set,
  from Paul Bolle.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-07 15:20:02 -05:00
David S. Miller
8b4cd8a053 Merge branch 'intel'
Jeff Kirsher says:

====================
This series contains updates to e1000e only.

All three patches come from Konstantin Khlebnikov to resolve power
management issues.  The first patch removes redundant and unbalanced
pci_disable_device() from the shutdown function.  The second patch
removes redundant actions from the driver and fixes the interaction
with actions in pci-bus runtime power management code.  The third
and last patch fixes some messages like 'Error reading PHY register'
and 'Hardware Erorr' and saves several seconds on reboot.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-07 15:18:53 -05:00
Heiko Carstens
0a75ca277c s390/kvm,gaccess: add address space annotations
Add missing address space annotations to all put_guest()/get_guest() callers.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 16:21:22 -03:00
Heiko Carstens
7c959e82ac s390/kvm: cleanup/fix handle_tpi()
- add missing specification exception check
- remove one level of indentation
- use defines instead of magic numbers

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Reviewed-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 16:21:22 -03:00
Heiko Carstens
f9dc72e82d s390/kvm,gaccess: shorten copy_to/from_guest code
The code can be significantly shortened. There is no functional change,
except that for large (> PAGE_SIZE) copies the guest translation would
be done more frequently.
However, there is not a single user which does this currently. If one
gets added later on this functionality can be added easily again.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 16:21:21 -03:00
Heiko Carstens
396083a964 s390/kvm,gaccess: shorten put/get_guest code
The put_guest_u*/get_guest_u* are nothing but wrappers for the regular
put_user/get_user uaccess functions. The only difference is that before
accessing user space the guest address must be translated to a user space
address.
Change the order of arguments for the guest access functions so they
match their uaccess parts. Also remove the u* suffix, so we simply
have put_guest/get_guest which will automatically use the right size
dependent on pointer type of the destination/source that now must be
correct.
In result the same behaviour as put_user/get_user except that accesses
must be aligned.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 16:21:21 -03:00
Heiko Carstens
dc5008b9bf s390/kvm: remove explicit -EFAULT return code checking on guest access
Let's change to the paradigm that every return code from guest memory
access functions that is not zero translates to -EFAULT and do not
explictly compare.
Explictly comparing the return value with -EFAULT has already shown to
be a bit fragile. In addition this is closer to the handling of
copy_to/from_user functions, which imho is in general a good idea.

Also shorten the return code handling in interrupt.c a bit.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 16:21:21 -03:00
Heiko Carstens
59a1fa2d80 s390/kvm,tprot: use new gmap_translate() function
When out-of-memory the tprot code incorrectly injected a program check
for the guest which reported an addressing exception even if the guest
address was valid.
Let's use the new gmap_translate() which translates a guest address to
a user space address whithout the chance of running into an out-of-memory
situation.
Also make it more explicit that for -EFAULT we won't find a vma.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 16:21:20 -03:00
Heiko Carstens
9e0fdb4145 s390/mm,gmap: implement gmap_translate()
Implement gmap_translate() function which translates a guest absolute address
to a user space process address without establishing the guest page table
entries.

This is useful for kvm guest address translations where no memory access
is expected to happen soon (e.g. tprot exception handler).

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 16:21:20 -03:00
Heiko Carstens
744b37fb5a s390/kvm,gaccess: fix guest access return code handling
Guest access functions like copy_to/from_guest() call __guestaddr_to_user()
which in turn call gmap_fault() in order to translate a guest address to a
user space address.
In error case __guest_addr_to_user() returns either -EFAULT or -ENOMEM.
The copy_to/from_guest functions just pass these return values down to the
callers.
The -ENOMEM case however is problematic since there are several places
which access guest memory like:

rc = copy_to_guest(...);
if (rc == -EFAULT)
	error_handling();

So in case of -ENOMEM the code assumes that the guest memory access
succeeded even though it failed.
This can cause guest data or state corruption.

If __guestaddr_to_user() returns -ENOMEM the meaning is that a valid user
space mapping exists, but there was not enough memory available when trying
to build the guest mapping. In other words an out-of-memory situation
occured.
For normal user space accesses an out-of-memory situation causes the page
fault handler to map -ENOMEM to -EFAULT (see fixup code in do_no_context()).
We need to do exactly the same for the kvm gaccess functions.

So __guestaddr_to_user() should just map all error codes to -EFAULT.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 16:21:19 -03:00
Jan Kiszka
1a0d74e664 KVM: nVMX: Fix setting of CR0 and CR4 in guest mode
The logic for calculating the value with which we call kvm_set_cr0/4 was
broken (will definitely be visible with nested unrestricted guest mode
support). Also, we performed the check regarding CR0_ALWAYSON too early
when in guest mode.

What really needs to be done on both CR0 and CR4 is to mask out L1-owned
bits and merge them in from L1's guest_cr0/4. In contrast, arch.cr0/4
and arch.cr0/4_guest_owned_bits contain the mangled L0+L1 state and,
thus, are not suited as input.

For both CRs, we can then apply the check against VMXON_CRx_ALWAYSON and
refuse the update if it fails. To be fully consistent, we implement this
check now also for CR4. For CR4, we move the check into vmx_set_cr4
while we keep it in handle_set_cr0. This is because the CR0 checks for
vmxon vs. guest mode will diverge soon when adding unrestricted guest
mode support.

Finally, we have to set the shadow to the value L2 wanted to write
originally.

Reviewed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 15:48:47 -03:00
Jan Kiszka
33fb20c39e KVM: nVMX: Fix content of MSR_IA32_VMX_ENTRY/EXIT_CTLS
Properly set those bits to 1 that the spec demands in case bit 55 of
VMX_BASIC is 0 - like in our case.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-03-07 15:47:11 -03:00
Dave Chinner
9e5987a779 xfs: rearrange some code in xfs_bmap for better locality
xfs_bmap.c is a big file, and some of the related code is spread all
throughout the file requiring function prototypes for static
function and jumping all through the file to follow a single call
path. Rearrange the code so that:

	a) related functionality is grouped together; and
	b) functions are grouped in call dependency order

While the diffstat is large, there are no code changes in the patch;
it is just moving the functionality around and removing the function
prototypes at the top of the file. The resulting layout of the code
is as follows (top of file to bottom):

	- miscellaneous helper functions
	- extent tree block counting routines
	- debug/sanity checking code
	- bmap free list manipulation functions
	- inode fork format manipulation functions
	- internal/external extent tree seach functions
	- extent tree manipulation functions used during allocation
	- functions used during extent read/allocate/removal
	  operations (i.e. xfs_bmapi_write, xfs_bmapi_read,
	  xfs_bunmapi and xfs_getbmap)

This means that following logic paths through the bmapi code is much
simpler - most of the code relevant to a specific operation is now
clustered together rather than spread all over the file....

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Mark Tinguely <tinguely@sgi.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
2013-03-07 12:35:22 -06:00
Akinobu Mita
ecb3403de1 xfs: rename random32() to prandom_u32()
Use more preferable function name which implies using a pseudo-random
number generator.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Acked-by: <bpm@sgi.com>
Cc: Ben Myers <bpm@sgi.com>
Cc: Alex Elder <elder@kernel.org>
Cc: xfs@oss.sgi.com
Signed-off-by: Ben Myers <bpm@sgi.com>
2013-03-07 12:33:57 -06:00
Dave Chinner
d5929de833 xfs: don't verify buffers after IO errors
When we read a buffer, we might get an error from the underlying
block device and not the real data. Hence if we get an IO error, we
shouldn't run the verifier but instead just pass the IO error
straight through.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Mark Tinguely <tinguely@sgi.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
2013-03-07 12:31:02 -06:00
Mark Tinguely
e8108cedb1 xfs: fix xfs_iomap_eof_prealloc_initial_size type
Fix the return type of xfs_iomap_eof_prealloc_initial_size() to
xfs_fsblock_t to reflect the fact that the return value may be an
unsigned 64 bits if XFS_BIG_BLKNOS is defined.

Signed-off-by: Mark Tinguely <tinguely@sgi.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
2013-03-07 12:29:59 -06:00
Brian Foster
e114b5fce6 xfs: increase prealloc size to double that of the previous extent
The updated speculative preallocation algorithm for handling sparse
files can becomes less effective in situations with a high number of
concurrent, sequential writers. The number of writers and amount of
available RAM affect the writeback bandwidth slicing algorithm,
which in turn affects the block allocation pattern of XFS. For
example, running 32 sequential writers on a system with 32GB RAM,
preallocs become fixed at a value of around 128MB (instead of
steadily increasing to the 8GB maximum as sequential writes
proceed).

Update the speculative prealloc heuristic to base the size of the
next prealloc on double the size of the preceding extent. This
preserves the original aggressive speculative preallocation
behavior and continues to accomodate sparse files at a slight cost
of increasing the size of preallocated data regions following holes
of sparse files.

Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
2013-03-07 12:28:25 -06:00