thr3ads.net - Xen devel - [PATCH v5 0/8] Relocate devices rather than memory for qemu-xen [Jun 2013]

If this information is useful, please help other people find it:
Share via:

George Dunlap

2013-Jun-21 16:08 UTC

[PATCH v5 0/8] Relocate devices rather than memory for qemu-xen

This is the fifth version of a patch series to address the issue of
qemu-xen not being able to handle moving guest memory in order to
resize the lowmem MMIO hole.

The only changes from v4 have to do with adding and correcting
comments.

A brief summary can be seen below:

A  1/8 hvmloader: Remove all 64-bit print arguments
A  2/8 hvmloader: Make the printfs more informative
A  3/8 hvmloader: Set up highmem resouce appropriately if there is no RAM above
4G
A  4/8 hvmloader: Fix check for needing a 64-bit bar
A- 5/8 hvmloader: Correct bug in low mmio region accounting
A  6/8 hvmloader: Load large devices into high MMIO space as needed
A  7/8 hvmloader: Remove minimum size for BARs to relocate to 64-bit space
a- 8/8 libxl,hvmloader: Don''t relocate memory for MMIO hole

Key
 -: Changes in v5 
 a: Reviewed / Acked by 1 person
 A: Reviewed / Acked by 2 or more people

= The situation = 

The default MMIO hole for Xen systems starts at 0xf0000000; this
leaves just under 256MiB of space for PCI devices.  At the moment,
hvmloader will scan the pci bus for devices and resize this hole, up
to 2GiB (i.e., starting at 0x80000000) to make space, relocating any
overlapping guest memory above 4GiB (0x100000000).  (After that point,
if there is still not enough space, the intention seemed to be that it
would begin mapping devices with 64-bit-capabile BARs into high memory
as well, just above the end of RAM; however, there seems to be a bug
in the code which detects this condition; it is likely that the 64-bit
remapping code was never capable of being triggered.)

We expect the default MMIO hole to be insufficient only when passing
through devices to guests.

This works fine for qemu-traditional, but qemu-xen unfortunately has
expectations of where guest memory will be, and will get confused if
it moves.  If hvmloader does relocate guest RAM, then at some point
qemu will try to map that pfn space, resulting in a seg fault and qemu
crashing.

hvmloader of course will only move RAM if it would overlap the MMIO
region; this means that if the guest has a small enough amount of RAM
-- say, 2GiB -- then this "memory move" condition will also never be
triggered. 

So at the moment, then under the following conditions:
 * A user is passing through a device or set of devices requiring more
 MMIO space than the default MMIO hole
 * The user has enough memory that resizing the hole will overlap
 guest memory
 * The user is using qemu-xen (not qemu-traditoinal)
then the user will shortly after boot experience qemu crashing.

= The proposed fix = 

This patch series makes the following functional changes:

The core change is this:

 * When running qemu-xen, don''t resize the MMIO hole; instead rely on
devices being moved into the 64-bit MMIO region.

In order to make this more effective, we also make the following
changes to the 64-bit relocation code:

 * Allow devices smaller than 512MiB to be relocated to the high MMIO
region.

 * When moving devices into the 64-bit MMIO region, start with the
ones with the largest BARs, and only relocate them if there is not
enough space for all the remaining BARs

= Risk analysis 
There are two kinds of risks: risks due to unintended changes (i.e., a
bug in the patch itself), and risks due to intended changes.

We hope that we can solve the first by a combination of testing and
code review.

The rest of this analysis will assume that the patch is correct, and
will try to do a risk analysis on the effects of the patch.

The main risk is that moving some devices into 64-bit memory will
cause problems with the operation of those devices.  Relocating a
device may have the following outcomes:
 1. In the best case, the relocated device will Just Work.  
 2. A relocated device may fail in a way that leaves the OS intact: the
guest OS may not be able to see them, or the driver may not load.
 3. A relocated device may fail in a way that crashes the guest OS: the
driver may crash, or one of the relocated devices which fails may be
system-critical.
 4. A relocated device may fail in a way which is unpredictable, but
does not cause data loss: crashing the guest randomly at some point in
the future, or causing strange quirks in functionality (e.g.,
network connectivity dropping, glitches when watching video).
 5. A relocated device may fail in a way that is unpredictable, and
corrupts data.

Outcomes 1-3 are equivalent or strictly better than crashing within a
few minutes of boot.  Outcome 4 is arguably also not much worse.

The main risk to our users would be #5.  However:
 - This is definitely a bug in the driver, OS, or the hardware
 - This is a bug that might be seen running on real hardware, or in
KVM (or some other hypervisor)
 - This is not a bug that we would be likely to catch, even if we had
a full development cycle worth of testing.

I think we should therefore not worry about #5, and consider in
general that relocating a device into 64-bit space will be no worse,
and potentially better, than crashing within a few minutes of boot.

There is another risk with this method, which is that a user may end
up passing through a number of devices with NON-64-bit BARs such that
the devices cannot all fit in the default lowmem MMIO region, but also
cannot be remapped above the 64-bit region.  If this is the case, then
some devices will simply not be able to be mapped.  If these
non-mapped devices are system critical, the VM will not boot; if they
are not, then the devices will simply be invisible.  Both of these are
either no worse than, and potentially better than, crashing within a
few minutes of boot.

Starting with all VMs:

Any VM running in PV mode will be unaffected.

Any VM running in HVM mode but not passing through devices will be
unaffected.

Any VM running in HVM mode and passing through devices that fit inside
the default MMIO space will be unaffected.

Any VM running in HVM mode, and passing through devices that require
less than 2GiB of MMIO space, *and* having a low enough guest memory
that the MMIO hole can be enlarged without moving guest memory, will
be unaffected.  (For example, if you need 512MiB and you have <3584
MiB of guest RAM; or if you need 1024MiB and have <3072 MiB of guest
RAM.)

Any VM running in HVM mode, is passing through devices requiring less
than 2GiB of MMIO space, and is using qemu-traditional will be
unaffected.

For a VM running in HVM mode, passing through devices which require more
than 2GiB of MMIO space, and using qemu-traditional, and having more
than 2GiB of guest memory:  
 * We believe that at the moment what will happen is that because of a
bug in hvmloader (fixed in this series), no devices will be mapped in
64-bit space; instead, the smallest devices will simply not be mapped.
This will likely cause critical platform devices not to be mapped,
causing the VM not to be able to boot.
 * With this patch, the largest devices *will* be remapped into 64-bit
space.  
 * If we are right that the current code will fail, this is a uniform
improvement, even if the devices don''t work.  
 * If the current code would work, then a different set of devices
will be re-mapped to high memory.  This may change some configurations
from "works" into "doesn''t work".

I think this is a small enough contingent of users, that this is an
acceptable amount of risk to take.

We have now covered all configurations of qemu-traditional.  Since
xend only knows how to use qemu-traditional, this also covers all
configurations using xend.

For VMs running in HVM mode, using qemu-xen, but not using libxl, this
patch will have no effect: qemu-xen will crash.  NB that this cannot
include xend, as it only knows how to drive qemu-traditional.  This
can be worked around by using qemu-traditional instead, or by setting
the appropriate xenstore key on boot.  This is acceptable, because
this is not really a supported configuration; users should use one of
the supported toolstacks, or use libxl.

We have now covered all users of any non-libxl-based toolstack.

For VM running in HVM mode, using qemu-xen, using libxl, and passing
through devices such that the required 32-bit only MMIO space does not
fit in the default MMIO hole, and with enough memory that resizing the
MMIO hole requires moving guest RAM:
 * At the moment, hvmloader will relocate guest memory.  This will
cause qemu-xen to crash within a few minutes.
 * With this change, the devices with the smallest BARs will simply
not be mapped.  If these devices are non-critical, they will simply be
invisible to the OS; if these devices are critical, the OS will not
boot.

Crashing immediately or having non-visible devices are the same or
better than crashing a few minutes into boot, so this is an
improvement (or at least not a regression).

For a VM running in HVM mode, using qemu-xen, using libxl, having a
required 32-bit only MMIO space that does fit within the default MMIO
hole, but a total MMIO space that does not, and having enough memory
that resizing the MMIO hole requires moving guest RAM:
 * At the moment, hvmloader will relocate memory.  This will cause
qemu-xen to crash within a few minutes of booting.  Note that this is
true whether the total MMIO space is less than 2GiB or more.
 * With this change, devices with the largest BARs will be relocated
to 64-bit space.  We expect that in general, the devices thus
relocated will be the passed-through PCI devices.

We have decided already to consider any outcome of mapping a device
into a 64-bit address space to be no worse than, and potentially
better than, qemu-xen crashing; so this can be considered an improvement.

We have now covered all possible configurations.

In summary:
 * The vast majority of configurations are unaffected
 * For those that are affected, the vast majority are either a strict
improvement, or no worse than, the status quo.
 * There is a slight possibility that in one extreme corner case
(using qemu-traditional with >2GiB of MMIO space), we may possibly be
changing "works" into "fails".  I think this is an
acceptable risk.

Therefore, I think the risks posed by this change are acceptable.

CC: George Dunlap <george.dunlap@eu.citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Ian Jackson <ian.jackson@citrix.com>
CC: Stefano Stabellini <stefano.stabellini@citrix.com>
CC: Hanweidong <hanweidong@huawei.com>
CC: Keir Fraser <keir@xen.org>

George Dunlap

2013-Jun-21 16:08 UTC

head link

[PATCH v5 1/8] hvmloader: Remove all 64-bit print arguments

The printf() available to hvmloader does not handle 64-bit data types;
manually break them down as two 32-bit strings.

v4:
 - Make macros for the requisite format and bit shifting

Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Keir Fraser <keir@xen.org>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Ian Jackson <ian.jackson@citrix.com>
CC: Stefano Stabellini <stefano.stabellini@citrix.com>
CC: Hanweidong <hanweidong@huawei.com>
CC: Keir Fraser <keir@xen.org>
---
 tools/firmware/hvmloader/pci.c  |   11 +++++++----
 tools/firmware/hvmloader/util.h |    2 ++
 2 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/tools/firmware/hvmloader/pci.c b/tools/firmware/hvmloader/pci.c
index c78d4d3..c1cb1e9 100644
--- a/tools/firmware/hvmloader/pci.c
+++ b/tools/firmware/hvmloader/pci.c
@@ -290,8 +290,9 @@ void pci_setup(void)
 
         if ( (base < resource->base) || (base > resource->max) )
         {
-            printf("pci dev %02x:%x bar %02x size %llx: no space for
"
-                   "resource!\n", devfn>>3, devfn&7,
bar_reg, bar_sz);
+            printf("pci dev %02x:%x bar %02x size "PRIllx": no
space for "
+                   "resource!\n", devfn>>3, devfn&7,
bar_reg,
+                   PRIllx_arg(bar_sz));
             continue;
         }
 
@@ -300,8 +301,10 @@ void pci_setup(void)
         pci_writel(devfn, bar_reg, bar_data);
         if (using_64bar)
             pci_writel(devfn, bar_reg + 4, bar_data_upper);
-        printf("pci dev %02x:%x bar %02x size %llx: %08x\n",
-               devfn>>3, devfn&7, bar_reg, bar_sz, bar_data);
+        printf("pci dev %02x:%x bar %02x size "PRIllx":
%08x\n",
+               devfn>>3, devfn&7, bar_reg,
+               PRIllx_arg(bar_sz),
+               bar_data);
 			
 
         /* Now enable the memory or I/O mapping. */
diff --git a/tools/firmware/hvmloader/util.h b/tools/firmware/hvmloader/util.h
index 7913259..9ccb905 100644
--- a/tools/firmware/hvmloader/util.h
+++ b/tools/firmware/hvmloader/util.h
@@ -168,6 +168,8 @@ void byte_to_hex(char *digits, uint8_t byte);
 void uuid_to_string(char *dest, uint8_t *uuid);
 
 /* Debug output */
+#define PRIllx "%x%08x"
+#define PRIllx_arg(ll) (uint32_t)((ll)>>32), (uint32_t)(ll)
 int printf(const char *fmt, ...) __attribute__ ((format (printf, 1, 2)));
 int vprintf(const char *fmt, va_list ap);
 
-- 
1.7.9.5

George Dunlap

2013-Jun-21 16:08 UTC

head link

[PATCH v5 2/8] hvmloader: Make the printfs more informative

* Warn that you''re relocating some BARs to 64-bit

* Warn that you''re relocating guest pages, and how many

* Include upper 32-bits of the base register when printing the bar
  placement info

v4:
 - Move message about relocating guest pages into loop, include number
   of pages and guest paddr
 - Fixed minor brace style issue

Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Ian Jackson <ian.jackson@citrix.com>
CC: Stefano Stabellini <stefano.stabellini@citrix.com>
CC: Hanweidong <hanweidong@huawei.com>
CC: Keir Fraser <keir@xen.org>
---
 tools/firmware/hvmloader/pci.c |   13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/tools/firmware/hvmloader/pci.c b/tools/firmware/hvmloader/pci.c
index c1cb1e9..44168e2 100644
--- a/tools/firmware/hvmloader/pci.c
+++ b/tools/firmware/hvmloader/pci.c
@@ -214,7 +214,11 @@ void pci_setup(void)
         pci_mem_start <<= 1;
 
     if ( (pci_mem_start << 1) != 0 )
+    {
+        printf("Low MMIO hole not large enough for all devices,"
+               " relocating some BARs to 64-bit\n");
         bar64_relocate = 1;
+    }
 
     /* Relocate RAM that overlaps PCI space (in 64k-page chunks). */
     while ( (pci_mem_start >> PAGE_SHIFT) < hvm_info->low_mem_pgend
)
@@ -227,6 +231,11 @@ void pci_setup(void)
         if ( hvm_info->high_mem_pgend == 0 )
             hvm_info->high_mem_pgend = 1ull << (32 - PAGE_SHIFT);
         hvm_info->low_mem_pgend -= nr_pages;
+        printf("Relocating 0x%x pages from "PRIllx" to
"PRIllx\
+               " for lowmem MMIO hole\n",
+               nr_pages,
+              
PRIllx_arg(((uint64_t)hvm_info->low_mem_pgend)<<PAGE_SHIFT),
+              
PRIllx_arg(((uint64_t)hvm_info->high_mem_pgend)<<PAGE_SHIFT));
         xatp.domid = DOMID_SELF;
         xatp.space = XENMAPSPACE_gmfn_range;
         xatp.idx   = hvm_info->low_mem_pgend;
@@ -301,10 +310,10 @@ void pci_setup(void)
         pci_writel(devfn, bar_reg, bar_data);
         if (using_64bar)
             pci_writel(devfn, bar_reg + 4, bar_data_upper);
-        printf("pci dev %02x:%x bar %02x size "PRIllx":
%08x\n",
+        printf("pci dev %02x:%x bar %02x size "PRIllx":
%x%08x\n",
                devfn>>3, devfn&7, bar_reg,
                PRIllx_arg(bar_sz),
-               bar_data);
+               bar_data_upper, bar_data);
 			
 
         /* Now enable the memory or I/O mapping. */
-- 
1.7.9.5

George Dunlap

2013-Jun-21 16:08 UTC

head link

[PATCH v5 3/8] hvmloader: Set up highmem resouce appropriately if there is no RAM above 4G

hvmloader will read hvm_info->high_mem_pgend to calculate where to
start the highmem PCI region.  However, if the guest does not have any
memory in the high region, this is set to zero, which will cause
hvmloader to use the "0" for the base of the highmem region, rather
than 1 << 32.

Check to see whether hvm_info->high_mem_pgend is set; if so, do the
normal calculation; otherwise, use 1<<32.

v4:
 - Handle case where hfm_info->high_mem_pgend is non-zero but
doesn''t
   point into high memory, throwing a warning.

Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Ian Jackson <ian.jackson@citrix.com>
CC: Stefano Stabellini <stefano.stabellini@citrix.com>
CC: Hanweidong <hanweidong@huawei.com>
CC: Keir Fraser <keir@xen.org>
---
 tools/firmware/hvmloader/pci.c |   13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/tools/firmware/hvmloader/pci.c b/tools/firmware/hvmloader/pci.c
index 44168e2..a3d03ed 100644
--- a/tools/firmware/hvmloader/pci.c
+++ b/tools/firmware/hvmloader/pci.c
@@ -246,7 +246,18 @@ void pci_setup(void)
         hvm_info->high_mem_pgend += nr_pages;
     }
 
-    high_mem_resource.base = ((uint64_t)hvm_info->high_mem_pgend) <<
PAGE_SHIFT;
+    high_mem_resource.base = ((uint64_t)hvm_info->high_mem_pgend) <<
PAGE_SHIFT;
+    if ( high_mem_resource.base < 1ull << 32 )
+    {
+        if ( hvm_info->high_mem_pgend != 0 )
+            printf("WARNING: hvm_info->high_mem_pgend %x"
+                   " does not point into high memory!",
+                   hvm_info->high_mem_pgend);
+        high_mem_resource.base = 1ull << 32;
+    }
+    printf("%sRAM in high memory; setting high_mem resource base to
"PRIllx"\n",
+           hvm_info->high_mem_pgend?"":"No ",
+           PRIllx_arg(high_mem_resource.base));
     high_mem_resource.max = 1ull << cpu_phys_addr();
     mem_resource.base = pci_mem_start;
     mem_resource.max = pci_mem_end;
-- 
1.7.9.5

George Dunlap

2013-Jun-21 16:08 UTC

head link

[PATCH v5 4/8] hvmloader: Fix check for needing a 64-bit bar

After attempting to resize the MMIO hole, the check to determine
whether there is a need to relocate BARs into 64-bit space checks the
specific thing that caused the loop to exit (MMIO hole == 2GiB) rather
than checking whether the required MMIO will fit in the hole.

But even then it does it wrong: the polarity of the check is
backwards.

Check for the actual condition we care about (the sizeof the MMIO
hole) rather than checking for the loop exit condition.

v3:
 - Move earlier in the series, before other functional changes

Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
CC: Ian Jackson <ian.jackson@citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Stefano Stabellini <stefano.stabellini@citrix.com>
CC: Hanweidong <hanweidong@huawei.com>
CC: Keir Fraser <keir@xen.org>
---
 tools/firmware/hvmloader/pci.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/firmware/hvmloader/pci.c b/tools/firmware/hvmloader/pci.c
index a3d03ed..6792ed4 100644
--- a/tools/firmware/hvmloader/pci.c
+++ b/tools/firmware/hvmloader/pci.c
@@ -213,7 +213,7 @@ void pci_setup(void)
             ((pci_mem_start << 1) != 0) )
         pci_mem_start <<= 1;
 
-    if ( (pci_mem_start << 1) != 0 )
+    if ( mmio_total > (pci_mem_end - pci_mem_start) )
     {
         printf("Low MMIO hole not large enough for all devices,"
                " relocating some BARs to 64-bit\n");
-- 
1.7.9.5

George Dunlap

2013-Jun-21 16:08 UTC

head link

[PATCH v5 5/8] hvmloader: Correct bug in low mmio region accounting

When deciding whether to map a device in low MMIO space (<4GiB),
hvmloader compares it with "mmio_left", which is set to the size of
the low MMIO range (pci_mem_end - pci_mem_start).  However, even if it
does map a device in high MMIO space, it still removes the size of its
BAR from mmio_left.

In reality we don''t need to do a separate accounting of the low memory
available -- this can be calculated from mem_resource.  Just get rid
of the variable and the duplicate accounting entirely.  This will make
the code more robust.

Note also that the calculation of whether to move a device to 64-bit
is fragile at the moment, depending on some unstated assumptions.
State those assumptions in a comment for future reference.

v5:
 - Add comment documenting fragility of the move-to-highmem check
v3:
 - Use mem_resource values directly instead of doing duplicate
   accounting

Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Stefano Stabellini <stefano.stabellini@citrix.com>
CC: Hanweidong <hanweidong@huawei.com>
CC: Keir Fraser <keir@xen.org>
---
 tools/firmware/hvmloader/pci.c |   20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)

diff --git a/tools/firmware/hvmloader/pci.c b/tools/firmware/hvmloader/pci.c
index 6792ed4..1fe250d 100644
--- a/tools/firmware/hvmloader/pci.c
+++ b/tools/firmware/hvmloader/pci.c
@@ -42,7 +42,6 @@ void pci_setup(void)
     uint32_t vga_devfn = 256;
     uint16_t class, vendor_id, device_id;
     unsigned int bar, pin, link, isa_irq;
-    int64_t mmio_left;
 
     /* Resources assignable to PCI devices via BARs. */
     struct resource {
@@ -264,8 +263,6 @@ void pci_setup(void)
     io_resource.base = 0xc000;
     io_resource.max = 0x10000;
 
-    mmio_left = pci_mem_end - pci_mem_start;
-
     /* Assign iomem and ioport resources in descending order of size. */
     for ( i = 0; i < nr_bars; i++ )
     {
@@ -273,7 +270,21 @@ void pci_setup(void)
         bar_reg = bars[i].bar_reg;
         bar_sz  = bars[i].bar_sz;
 
-        using_64bar = bars[i].is_64bar && bar64_relocate &&
(mmio_left < bar_sz);
+        /*
+         * NB: The code here is rather fragile, as the check here to see
+         * whether bar_sz will fit in the low MMIO region doesn''t
match the
+         * real check made below, which involves aligning the base offset of
the
+         * bar with the size of the bar itself.  As it happens, this will
always
+         * be satisfied because:
+         * - The first one will succeed because the MMIO hole can only start at
+         *   0x{f,e,c,8}00000000.  If it fits, it will be aligned properly.
+         * - All subsequent ones will be aligned because the list is ordered
+         *   large to small, and bar_sz is always a power of 2. (At least
+         *   the code here assumes it to be.)
+         * Should either of those two conditions change, this code will break.
+         */
+        using_64bar = bars[i].is_64bar && bar64_relocate
+            && (bar_sz > (mem_resource.max - mem_resource.base));
         bar_data = pci_readl(devfn, bar_reg);
 
         if ( (bar_data & PCI_BASE_ADDRESS_SPACE) =@@ -295,7 +306,6 @@ void
pci_setup(void)
                 resource = &mem_resource;
                 bar_data &= ~PCI_BASE_ADDRESS_MEM_MASK;
             }
-            mmio_left -= bar_sz;
         }
         else
         {
-- 
1.7.9.5

George Dunlap

2013-Jun-21 16:08 UTC

head link

[PATCH v5 6/8] hvmloader: Load large devices into high MMIO space as needed

Keep track of how much mmio space is left total, as well as the amount
of "low" MMIO space (<4GiB), and only load devices into high memory
if
there is not enough low memory for the rest of the devices to fit.

Because devices are processed by size in order from large to small,
this should preferentially relocate devices with large BARs to 64-bit
space.

v3:
 - Just use mmio_total rather than introducing a new variable.
 - Port to using mem_resource directly rather than low_mmio_left

Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Stefano Stabellini <stefano.stabellini@citrix.com>
CC: Hanweidong <hanweidong@huawei.com>
CC: Keir Fraser <keir@xen.org>
---
 tools/firmware/hvmloader/pci.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/tools/firmware/hvmloader/pci.c b/tools/firmware/hvmloader/pci.c
index 1fe250d..68d4e44 100644
--- a/tools/firmware/hvmloader/pci.c
+++ b/tools/firmware/hvmloader/pci.c
@@ -271,6 +271,11 @@ void pci_setup(void)
         bar_sz  = bars[i].bar_sz;
 
         /*
+         * Relocate to high memory if the total amount of MMIO needed
+         * is more than the low MMIO available.  Because devices are
+         * processed in order of bar_sz, this will preferentially
+         * relocate larger devices to high memory first.
+         *
          * NB: The code here is rather fragile, as the check here to see
          * whether bar_sz will fit in the low MMIO region doesn''t
match the
          * real check made below, which involves aligning the base offset of
the
@@ -284,7 +289,7 @@ void pci_setup(void)
          * Should either of those two conditions change, this code will break.
          */
         using_64bar = bars[i].is_64bar && bar64_relocate
-            && (bar_sz > (mem_resource.max - mem_resource.base));
+            && (mmio_total > (mem_resource.max -
mem_resource.base));
         bar_data = pci_readl(devfn, bar_reg);
 
         if ( (bar_data & PCI_BASE_ADDRESS_SPACE) =@@ -306,6 +311,7 @@ void
pci_setup(void)
                 resource = &mem_resource;
                 bar_data &= ~PCI_BASE_ADDRESS_MEM_MASK;
             }
+            mmio_total -= bar_sz;
         }
         else
         {
-- 
1.7.9.5

George Dunlap

2013-Jun-21 16:08 UTC

head link

[PATCH v5 7/8] hvmloader: Remove minimum size for BARs to relocate to 64-bit space

Allow devices with BARs less than 512MiB to be relocated to high
memory.

This will only be invoked if there is not enough low MMIO space to map
the device, and will be done preferentially to large devices first; so
in all likelihood only large devices will be remapped anyway.

This is needed to work-around the issue of qemu-xen not being able to
handle moving guest memory around to resize the MMIO hole.  The
default MMIO hole size is less than 256MiB.

v3:
 - Fixed minor style issue

Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Hanweidong <hanweidong@huawei.com>
CC: Keir Fraser <keir@xen.org>
---
 tools/firmware/hvmloader/config.h |    1 -
 tools/firmware/hvmloader/pci.c    |    5 ++---
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/tools/firmware/hvmloader/config.h
b/tools/firmware/hvmloader/config.h
index 8143d6f..6641197 100644
--- a/tools/firmware/hvmloader/config.h
+++ b/tools/firmware/hvmloader/config.h
@@ -55,7 +55,6 @@ extern struct bios_config ovmf_config;
 /* MMIO hole: Hardcoded defaults, which can be dynamically expanded. */
 #define PCI_MEM_START       0xf0000000
 #define PCI_MEM_END         0xfc000000
-#define PCI_MIN_BIG_BAR_SIZE          0x20000000
 
 extern unsigned long pci_mem_start, pci_mem_end;
 
diff --git a/tools/firmware/hvmloader/pci.c b/tools/firmware/hvmloader/pci.c
index 68d4e44..9253b0b 100644
--- a/tools/firmware/hvmloader/pci.c
+++ b/tools/firmware/hvmloader/pci.c
@@ -295,9 +295,8 @@ void pci_setup(void)
         if ( (bar_data & PCI_BASE_ADDRESS_SPACE) =             
PCI_BASE_ADDRESS_SPACE_MEMORY )
         {
-            /* Mapping high memory if PCI deivce is 64 bits bar and the bar
size
-               is larger than 512M */
-            if (using_64bar && (bar_sz > PCI_MIN_BIG_BAR_SIZE)) {
+            /* Mapping high memory if PCI device is 64 bits bar */
+            if ( using_64bar ) {
                 if ( high_mem_resource.base & (bar_sz - 1) )
                     high_mem_resource.base = high_mem_resource.base - 
                         (high_mem_resource.base & (bar_sz - 1)) + bar_sz;
-- 
1.7.9.5

George Dunlap

2013-Jun-21 16:08 UTC

head link

[PATCH v5 8/8] libxl, hvmloader: Don''t relocate memory for MMIO hole

At the moment, qemu-xen can''t handle memory being relocated by
hvmloader.  This may happen if a device with a large enough memory
region is passed through to the guest.  At the moment, if this
happens, then at some point in the future qemu will crash and the
domain will hang.  (qemu-traditional is fine.)

It''s too late in the release to do a proper fix, so we try to do
damage control.

hvmloader already has mechanisms to relocate memory to 64-bit space if
it can''t make a big enough MMIO hole.  By default this is 2GiB; if we
just refuse to make the hole bigger if it will overlap with guest
memory, then the relocation will happen by default.

v5:
 - Update comment to not refer to "this series".
v4:
 - Wrap long line in libxl_dm.c
 - Fix comment
v3:
 - Fix polarity of comparison
 - Move diagnostic messages to another patch
 - Tested with xen platform pci device hacked to have different BAR sizes
   {256MiB, 1GiB} x {qemu-xen, qemu-traditional} x various memory
   configurations
 - Add comment explaining why we default to "allow"
 - Remove cast to bool
v2:
 - style fixes
 - fix and expand comment on the MMIO hole loop
 - use "%d" rather than "%s" ->
(...)?"1":"0"
 - use bool instead of uint8_t
 - Move 64-bit bar relocate detection to another patch
 - Add more diagnostic messages

Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Stefano Stabellini <stefano.stabellini@citrix.com>
CC: Hanweidong <hanweidong@huawei.com>
CC: Keir Fraser <keir@xen.org>
CC: Keir Fraser <keir@xen.org>
---
 tools/firmware/hvmloader/pci.c          |   49 +++++++++++++++++++++++++++++--
 tools/libxl/libxl_dm.c                  |    8 +++++
 xen/include/public/hvm/hvm_xs_strings.h |    1 +
 3 files changed, 56 insertions(+), 2 deletions(-)

diff --git a/tools/firmware/hvmloader/pci.c b/tools/firmware/hvmloader/pci.c
index 9253b0b..627e8cb 100644
--- a/tools/firmware/hvmloader/pci.c
+++ b/tools/firmware/hvmloader/pci.c
@@ -27,6 +27,8 @@
 
 #include <xen/memory.h>
 #include <xen/hvm/ioreq.h>
+#include <xen/hvm/hvm_xs_strings.h>
+#include <stdbool.h>
 
 unsigned long pci_mem_start = PCI_MEM_START;
 unsigned long pci_mem_end = PCI_MEM_END;
@@ -57,6 +59,32 @@ void pci_setup(void)
     } *bars = (struct bars *)scratch_start;
     unsigned int i, nr_bars = 0;
 
+    const char *s;
+    /*
+     * Do we allow hvmloader to relocate guest memory in order to
+     * increase the size of the lowmem MMIO hole?  Defaulting to 1
+     * here will mean that non-libxl toolstacks (including xend and
+     * home-grown ones) means that those using qemu-xen will still
+     * experience the memory relocation bug described below; but it
+     * also means that those using qemu-traditional will *not*
+     * experience any change; and it also means that there is a
+     * work-around for those using qemu-xen, namely switching to
+     * qemu-traditional.
+     *
+     * If we defaulted to 0, and failing to resize the hole caused any
+     * problems with qemu-traditional, then there is no work-around.
+     *
+     * Since xend can only use qemu-traditional, I think this is the
+     * option that will have the least impact.
+     */
+    bool allow_memory_relocate = 1;
+
+    s = xenstore_read(HVM_XS_ALLOW_MEMORY_RELOCATE, NULL);
+    if ( s )
+        allow_memory_relocate = strtoll(s, NULL, 0);
+    printf("Relocating guest memory for lowmem MMIO space %s\n",
+           allow_memory_relocate?"enabled":"disabled");
+
     /* Program PCI-ISA bridge with appropriate link routes. */
     isa_irq = 0;
     for ( link = 0; link < 4; link++ )
@@ -208,8 +236,25 @@ void pci_setup(void)
         pci_writew(devfn, PCI_COMMAND, cmd);
     }
 
-    while ( (mmio_total > (pci_mem_end - pci_mem_start)) &&
-            ((pci_mem_start << 1) != 0) )
+    /*
+     * At the moment qemu-xen can''t deal with relocated memory
regions.
+     * It''s too close to the release to make a proper fix; for now,
+     * only allow the MMIO hole to grow large enough to move guest memory
+     * if we''re running qemu-traditional.  Items that don''t
fit will be
+     * relocated into the 64-bit address space.
+     *
+     * This loop now does the following:
+     * - If allow_memory_relocate, increase the MMIO hole until it''s
+     *   big enough, or until it''s 2GiB
+     * - If !allow_memory_relocate, increase the MMIO hole until it''s
+     *   big enough, or until it''s 2GiB, or until it overlaps guest
+     *   memory
+     */
+    while ( (mmio_total > (pci_mem_end - pci_mem_start)) 
+            && ((pci_mem_start << 1) != 0)
+            && (allow_memory_relocate
+                || (((pci_mem_start << 1) >> PAGE_SHIFT)
+                    >= hvm_info->low_mem_pgend)) )
         pci_mem_start <<= 1;
 
     if ( mmio_total > (pci_mem_end - pci_mem_start) )
diff --git a/tools/libxl/libxl_dm.c b/tools/libxl/libxl_dm.c
index ac1f90e..7e54c02 100644
--- a/tools/libxl/libxl_dm.c
+++ b/tools/libxl/libxl_dm.c
@@ -1154,6 +1154,14 @@ void libxl__spawn_local_dm(libxl__egc *egc,
libxl__dm_spawn_state *dmss)
         libxl__xs_write(gc, XBT_NULL,
                         libxl__sprintf(gc, "%s/hvmloader/bios",
path),
                         "%s",
libxl_bios_type_to_string(b_info->u.hvm.bios));
+        /* Disable relocating memory to make the MMIO hole larger
+         * unless we''re running qemu-traditional */
+        libxl__xs_write(gc, XBT_NULL,
+                        libxl__sprintf(gc,
+                                      
"%s/hvmloader/allow-memory-relocate",
+                                       path),
+                        "%d",
+                       
b_info->device_model_version==LIBXL_DEVICE_MODEL_VERSION_QEMU_XEN_TRADITIONAL);
         free(path);
     }
 
diff --git a/xen/include/public/hvm/hvm_xs_strings.h
b/xen/include/public/hvm/hvm_xs_strings.h
index 9042303..4de5881 100644
--- a/xen/include/public/hvm/hvm_xs_strings.h
+++ b/xen/include/public/hvm/hvm_xs_strings.h
@@ -28,6 +28,7 @@
 #define HVM_XS_HVMLOADER               "hvmloader"
 #define HVM_XS_BIOS                    "hvmloader/bios"
 #define HVM_XS_GENERATION_ID_ADDRESS  
"hvmloader/generation-id-address"
+#define HVM_XS_ALLOW_MEMORY_RELOCATE  
"hvmloader/allow-memory-relocate"
 
 /* The following values allow additional ACPI tables to be added to the
  * virtual ACPI BIOS that hvmloader constructs. The values specify the guest
-- 
1.7.9.5

Ian Jackson

2013-Jun-24 11:27 UTC

head link

Re: [PATCH v5 0/8] Relocate devices rather than memory for qemu-xen

George Dunlap writes ("[PATCH v5 0/8] Relocate devices rather than memory
for qemu-xen"):> This is the fifth version of a patch series to address the issue of
> qemu-xen not being able to handle moving guest memory in order to
> resize the lowmem MMIO hole.
> 
> The only changes from v4 have to do with adding and correcting
> comments.
I have applied this series, thanks.

Ian.

Ian Campbell

2013-Jun-25 13:23 UTC

head link

Re: [PATCH v5 8/8] libxl, hvmloader: Don''t relocate memory for MMIO hole

On Fri, 2013-06-21 at 17:08 +0100, George Dunlap wrote:> 
> diff --git a/xen/include/public/hvm/hvm_xs_strings.h
b/xen/include/public/hvm/hvm_xs_strings.h
> index 9042303..4de5881 100644
> --- a/xen/include/public/hvm/hvm_xs_strings.h
> +++ b/xen/include/public/hvm/hvm_xs_strings.h
> @@ -28,6 +28,7 @@
>  #define HVM_XS_HVMLOADER               "hvmloader"
>  #define HVM_XS_BIOS                    "hvmloader/bios"
>  #define HVM_XS_GENERATION_ID_ADDRESS  
"hvmloader/generation-id-address"
> +#define HVM_XS_ALLOW_MEMORY_RELOCATE  
"hvmloader/allow-memory-relocate"
Please could you also followup with a patch to
docs/misc/xenstore-paths.markdown describing this key.

Ian.

Xen devel - Jun 2013 - [PATCH v5 0/8] Relocate devices rather than memory for qemu-xen

[PATCH v5 0/8] Relocate devices rather than memory for qemu-xen

[PATCH v5 1/8] hvmloader: Remove all 64-bit print arguments

[PATCH v5 2/8] hvmloader: Make the printfs more informative

[PATCH v5 3/8] hvmloader: Set up highmem resouce appropriately if there is no RAM above 4G

[PATCH v5 4/8] hvmloader: Fix check for needing a 64-bit bar

[PATCH v5 5/8] hvmloader: Correct bug in low mmio region accounting

[PATCH v5 6/8] hvmloader: Load large devices into high MMIO space as needed

[PATCH v5 7/8] hvmloader: Remove minimum size for BARs to relocate to 64-bit space

[PATCH v5 8/8] libxl, hvmloader: Don''t relocate memory for MMIO hole

Re: [PATCH v5 0/8] Relocate devices rather than memory for qemu-xen

Re: [PATCH v5 8/8] libxl, hvmloader: Don''t relocate memory for MMIO hole