similar to: Overhead of each Hypercall

Sorry, forget to join lists. Regards, Jinsong -----Original Message----- From: Liu, Jinsong Sent: 2007年5月16日 8:23 To: ''Keir Fraser''; Tian, Kevin; xen-devel@lists.xensource.com Subject: RE: [Xen-devel] trap/interrupt gate for hypercall In native linux, syscall (gate type 15) and its RESTORE_ALL code provides weak fixup mechanism when reload segment register, by sending 0

On 04/30/2015 06:39 PM, Jeremy Fitzhardinge wrote: > On 04/30/2015 03:53 AM, Juergen Gross wrote: >> Paravirtualized spinlocks produce some overhead even if the kernel is >> running on bare metal. The main reason are the more complex locking >> and unlocking functions. Especially unlocking is no longer just one >> instruction but so complex that it is no longer inlined.

On 04/30/2015 06:39 PM, Jeremy Fitzhardinge wrote: > On 04/30/2015 03:53 AM, Juergen Gross wrote: >> Paravirtualized spinlocks produce some overhead even if the kernel is >> running on bare metal. The main reason are the more complex locking >> and unlocking functions. Especially unlocking is no longer just one >> instruction but so complex that it is no longer inlined.

Hi, I have *implemented a new hypercall* and it is working fine when tested from the Dom0 user-space. I want to invoke this hypercall from DomU user-space. I copied all the /usr/lib/libxen* and /usr/include/xen* (recursively) to the DomU Here''s the code I wrote to invoke hypercall: #include <stdio.h> #include <xenctrl.h> int main(void){ int xc_handle, rc;

* Juergen Gross <jgross at suse.com> wrote: > On 05/05/2015 07:21 PM, Jeremy Fitzhardinge wrote: > >On 05/03/2015 10:55 PM, Juergen Gross wrote: > >>I did a small measurement of the pure locking functions on bare metal > >>without and with my patches. > >> > >>spin_lock() for the first time (lock and code not in cache) dropped from >

Hi, I''d like to do some operation every time an hypercall is issued in a domU kernel. I started modifying include/asm-i386/mach-xen/asm/hypercall.h but I don''t think it''s such a great idea, so I think that, as a last chance, I''ll have to substitute all of the hypercall function call with some wrapper function of my own. Can anyone tell me if

For my debugging pleasure I have added another domctrl to call trace_var from xc_save. That works as expected. However, it adds additional pressure on the tracebuffer because each hypercall is traced as well. My question is: Would an additional domctrl (or whatever), or another trace hypercall be acceptable? Another hypercall would at least have a chance to be ignored by trace_hypercall()

Hi, I implemented my own Hypercall in xen. Now I wish my application under dom0 could talk to Xen via my hypercall. Currently, I make a stub in dom0''s kernel, which works as a proxy to forward the request. So, my application first makes a system call to dom0, which then makes hypercall to xen. Is this design necessary? May I call hypercalls from user application under dom0 directly?

I''m tinkering with some code to be able to talk to the hypervisor from Windows, with the ultimate aim to get a usable xenbus. So far I''ve created a driver for the Xen PCI ''device'', and (hopefully) mapped the hypercall space given by the CPUID call... What is the simplest hypercall I can make to get an answer from the hypervisor to tell me that it''s

This patch pretty prints the hypercall section for $xenperf -f Each hypercall count is tagged by its name. Reference: http://lists.xensource.com/archives/html/xen-ia64-devel/2006-08/msg00261.html Signed-off-by Ken Hironaka <kenny@logos.ic.i.u-tokyo.ac.jp> _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com

Hi all, I have been seeing a considerable (2x-3x) overhead when using Posix pthreads and pthreads mutux synchronization on Xen dom0. I am not sure if this is an inherent overhead of Xen or a wrong configuration on my side. Basically, I run a simple benchmark to time pthread_create and pthread_mutex_lock/unlock for 200 threads and the overhead I am measuring for xen dom0 is 1.5x-2.9x relative to

Hi, I am working on a security architecture. In this architecture, the application in DomU has to communicate directly with the hypervisor. But as I can see, the xen architecture allows only DomU kernel to raise a hypercall. I am planning to enable application to communicate with xen directly. I am assuming, setting up a trap gate with Ring-3 access should do the trick. I have few questions

Keir Fraser wrote: > On 19 Apr 2006, at 17:44, Hollis Blanchard wrote: > >> "xen_frameno_t" then? > > xen_pfn_t? Definitely won''t conflict with anyone, and I prefer ''pfn'' > to ''frameno'' as it''s more consistent with other names we have in the > interface. > >> Attached is the updated patch, with

For an SMP Linux HVM guest with PV drivers inserted, when we do save/restore (or LiveMigration) for the guest, it might panic after it''s restored. The panic point is inside ap_suspend(): .... while (info->do_spin) { cpu_relax(); read_lock(&suspend_lock); HYPERVISOR_yield(); ----> guest might panic on the invocation of this function.

Hi, the attached patch makes Xen tracing work again, after the introduction of the hypercall buffers broke it. Just a missing line. Thanks to Uwe Dannowski for reporting this. Signed-off-by: Andre Przywara <andre.przywara@amd.com> Regards, Andre. -- Andre Przywara AMD-Operating System Research Center (OSRC), Dresden, Germany Tel: +49 351 448-3567-12

What are the possible cases (I hope there are not many:)) that this Hypercall might fail, HYPERVISOR_event_channel_op(EVTCHNOP_send, &op)???,With an error value of "-ENOSYS (-38)" on a "HVM"? Thx, Venkat _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

Can anyone provide a code fragment of what a hypercall would look like under a HVM Windows domain? Or is the answer more complicated than the question assumes? Thanks James _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

xentrace, setsize and the hypervisor don''t seem to be on the same page. If tracing is not enabled, the hypercall will only allow TBUF_SET_SIZE and TBUF_ENABLE. However, set_size calls TBUF_GET_INFO, and xentrace wants to call TBUF_SET_EVT_MASK and TBUF_SET_CPU_MASK before calling TBUF_ENABLE. I don''t see any reason not to call SET_*_MASK and GET_INFO if tracing is not enabled.

Hi, I want to get the virtual addresses of the hypercall entry and exit. To know system call addresses, I know there is a file System.map in /boot. By looking up the file we can get many kernel symbols addresses. I am wondering if there is a similar file in Xen. Or is there other ways so that I can determine the virtual addresses of hypercalls. Any information is appreciated. Thanks in advance.

add a hypercall number for virtual device in unmodified guest There are 6 patches for para-driver support in vmx guest. This is the first one. Signed-off-by: Xiaofeng Ling <xiaofeng.ling@intel.com> Signed-off-by: Arun Sharma <arun.sharma@intel.com> diff -r 287d36b46fa3 xen/arch/x86/x86_32/entry.S --- a/xen/arch/x86/x86_32/entry.S Tue Aug 30 20:36:49 2005 +++