Use non-trivial dma_{un,}map_page only when CONFIG_HIGHMEM, allowing
to not compile swiotlb_{un,}map_page in that case, which in neither
case need exporting.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Use address width needed by device rather than dma_bits in
dma_alloc_coherent().
Probe supported address width in swiotlb initialization.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
While, as discussed, the performance impact of this option is
certainly higher than on native Linux, the option should not be
entirely disallowed if people want to sacrifice performance for less
lowmem pressure.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
PUDs,PMDs,PTEs are all marked as ForeignPage so that they can be
grabbed from tlb_remove_page() at the appropriate time and freed in a
special way.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
This causes the kernel to use a polling loop rather than hypercall
down to block. I have found this useful to profile the amount of sleep
cycles when using xenoprof, there may be other uses.
Signed-off-by: Amitabha Roy <amitabha.roy@gmail.com>
The 'll' length modifier is not available with scanf (and therefore
xenbus_gather) on older kernel versions so use 'L' which is available at
least back to 2.6.5.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
The recent change to use a default DMA bit width of 30 bits (required
by chips like the b44) only converted the streaming DMA primitives.
The coherent mappings are still hard-coded to 31 bits. This means
that b44 still doesn't work under Xen.
This patch makes the io_tlb_dma_bits variable global and uses it for
coherent memory mappings.
Thanks to Calvin Webster for providing a machine with a b44 and 2G
of memory I've been able to verify that this finally makes the b44
work under Xen.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Rename 'io_tlb_dma_bits' to the more correct 'dma_bits'. This also
affects the name of the boot parameter, which is now 'dma_bits='.
Signed-off-by: Keir Fraser <keir@xensource.com>
This eliminates the risk of overflowing the kernel stack and is a
reasonable policy given that we have no concept of priorities among
event sources.
Signed-off-by: Keir Fraser <keir@xensource.com>
Since we had changed to use event callback to deliver interrupts,
1. The pending_interruption is changed to pending_event.
2. get_ivr, set_tpr, get_trp and set_eoi are not used or only used
in the initialization phase. There is no need to write this code
in assembly. This code is deleted.
3. hyper_ssm_i needs to be rewritten to jump to entchn_callback_handler
instead of iva+0x3000 interrupt handler. I will do this later.
Signed-off-by: Anthony Xu <anthony.xu@intel.com>
Changeset 13018:c98ca86138a7422cdf9b15d87c95619b7277bb6a merely sweeps
the bug under the carpet: it silences zap_page_range()'s cries for
help by keeping interrupts enabled. That doesn't fix the bug, and
it's also wrong: if a critical region gets interrupted, and the
interrupt printk()s, xenfb_refresh() gets executed and promptly
deadlocks.
This patch fixes the locking, but leaves open a race between
xenfb_update_screen() and do_no_page(). See the source code for a
detailed explanation of how it works, and where it fails.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
skb_segment fails to segment linear packets correctly because it
tries to write all linear parts of the original skb into each
segment. This will always panic as each segment only contains
enough space for one MSS.
This was not detected earlier because linear packets should be
rare for GSO. In fact it still remains to be seen what exactly
created the linear packets that triggered this bug. Basically
the only time this should happen is if someone enables GSO
emulation on an interface that does not support SG.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rename xen_machine_kexec() to just machine_kexec() and only compile in
the Xen or native version as appropriate.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
x86 foreign HVM domain page mapping semantic was changed to use gmfn
instead mfn. It applies to domains with auto_translated_mode enabled,
and all ia64 domains enable auto_translated_mode. This patch changes
ia64 foreign domain page mapping to use gmfn and fixes ia64 domU buidler.
However this patch breaks domain save/restore/dump-core.
They should also be fixed-up
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
resource.
This was exposed by 12709:98413fa7826c which corrected the length of
the hypervisor resource.
Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
