search for: xs_unwatch

XenStore puts in queued watch events via a thread and notifies the user. Sometimes xs_unwatch is called before all related message is read. The use case is non-threaded libevent, we have two event A and B: - Event A will destroy something and call xs_unwatch; - Event B is used to notify that a node has changed in XenStore. As the event is called one by one, event A can be handled be...

XenStore puts in queued watch events via a thread and notifies the user. Sometimes xs_unwatch is called before all related message is read. The use case is non-threaded libevent, we have two event A and B: - Event A will destroy something and call xs_unwatch; - Event B is used to notify that a node has changed in XenStore. As the event is called one by one, event A can be handled be...

While on entry to xs_unwatch, there may be an event on its way from xenstored (eg in the ring or in the local kernel), all such events will definitely come before the reply to the unwatch command. So at the point where the unwatch reply has been processed (after xs_talkv), any such now-deleted watch events will definitely hav...

XenStore puts in queued watch events via a thread and notifies the user. Sometimes xs_unwatch is called before all related message is read. The use case is non-threaded libevent, we have two event A and B: - Event A will destroy something and call xs_unwatch; - Event B is used to notify that a node has changed in XenStore. As the event is called one by one, event A can be handled be...

XenStore puts in queued watch events via a thread and notifies the user. Sometimes xs_unwatch is called before all related message is read. The use case is non-threaded libevent, we have two event A and B: - Event A will destroy something and call xs_unwatch; - Event B is used to notify that a node has changed in XenStore. As the event is called one by one, event A can be handled be...

Hi, George sent a patch on this problem before, but it was not completed. This patch makes libxl use xenbus to communicate with xenstored if libxl cannot open a socket. There''s a place that does not close fd in the case of failure, which is also fixed in this patch. -----------------------------------------Patch------------------------------------------------------ diff -r eff592364826

...uint32_t msg_type; void *reply; + char *path, *token; + struct watch_adapter *watch, *tmp_watch; + int err; if ((len + u->len) > sizeof(u->u.buffer)) return -EINVAL; @@ -169,6 +223,56 @@ static ssize_t xenbus_dev_write(struct f kfree(reply); break; + case XS_WATCH: + case XS_UNWATCH: + path = u->u.buffer + sizeof(u->u.msg); + token = memchr(path, 0, u->u.msg.len); + if (token == NULL) + return -EILSEQ; + token++; + + if (msg_type == XS_WATCH) { + static const char * XS_WATCH_RESP = "OK"; + struct xsd_sockmsg hdr; + + watch = kmalloc(sizeof(*watc...

...00 @@ -293,12 +293,14 @@ static bool watch_domain(struct domain * bool success; sprintf(domid_str, "dom%u", dom->domid); - if (watch) + if (watch) { success = xs_watch(xs, dom->conspath, domid_str); - else + if (success) + domain_create_ring(dom); + } else { success = xs_unwatch(xs, dom->conspath, domid_str); - if (success) - domain_create_ring(dom); + } + return success; } _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel

Dear xen-devel, There''s a race condition in xenfs (xenstore driver) that causes userspace utility xenstore-watch to crash. Normally, the userspace process gets an "OK" from xenfs and then the watch fires immediately after. Occasionally, this happens the other way around: the watch fires before the driver sends "OK", which confuses the xenstore-watch client. It seems

...r* xenbus_unwatch_path_token( xenbus_transaction_t xbt, const char *path, const char *token) +{ + struct xsd_sockmsg *rep; + + struct write_req req[] = { + {path, strlen(path) + 1}, + {token, strlen(token) + 1}, + }; + + struct watch *watch, **prev; + + rep = xenbus_msg_reply(XS_UNWATCH, xbt, req, ARRAY_SIZE(req)); + + char *msg = errmsg(rep); + if (msg) return msg; + free(rep); + + for (prev = &watches, watch = *prev; watch; prev = &watch->next, watch = *prev) + if (!strcmp(watch->token, token)) { + free(watch->token); + *...

...ck(conn, XS_DEBUG); failtest = true; } - return false; - } #endif /* TESTING */ + break; case XS_WATCH: - return do_watch(conn, in); + do_watch(conn, in); + break; case XS_WATCH_ACK: - return do_watch_ack(conn, onearg(in)); + do_watch_ack(conn, onearg(in)); + break; case XS_UNWATCH: - return do_unwatch(conn, in); + do_unwatch(conn, in); + break; case XS_TRANSACTION_START: - return do_transaction_start(conn, onearg(in)); + do_transaction_start(conn, onearg(in)); + break; case XS_TRANSACTION_END: - return do_transaction_end(conn, onearg(in)); + do_transaction_en...

The following series of 10 patches is a merge of the xenfb and xenconsoled functionality into the qemu-dm code. The general approach taken is to have qemu-dm provide two machine types - one for xen paravirt, the other for fullyvirt. For compatability the later is the default. The goals overall are to kill LibVNCServer, remove alot of code duplication and/or parallel impls of the same concepts, and

...me_domain at VERS_4.0 4.16.0 - xs_rm at VERS_4.0 4.16.0 - xs_set_permissions at VERS_4.0 4.16.0 - xs_set_target at VERS_4.0 4.16.0 - xs_strings_to_perms at VERS_4.0 4.16.0 - xs_suspend_evtchn_port at VERS_4.0 4.16.0 - xs_transaction_end at VERS_4.0 4.16.0 - xs_transaction_start at VERS_4.0 4.16.0 - xs_unwatch at VERS_4.0 4.16.0 - xs_watch at VERS_4.0 4.16.0 - xs_write at VERS_4.0 4.16.0 - xs_write_all at VERS_4.0 4.16.0 diff -Nru xen-4.17.3+10-g091466ba55/debian/libxenstore4t64.install xen-4.17.3+10-g091466ba55/debian/libxenstore4t64.install --- xen-4.17.3+10-g091466ba55/debian/libxenstore4t64.install 1...

...{ l1 = xs_read_watch(ctx->xsh, &n); if (l1 != NULL) { - char *state = libxl_xs_read(ctx, XBT_NULL, l1[XS_WATCH_PATH]); + char *state = libxl_xs_read(&gc, XBT_NULL, l1[XS_WATCH_PATH]); if (!state || atoi(state) == 6) { xs_unwatch(ctx->xsh, l1[0], l1[1]); xs_rm(ctx->xsh, XBT_NULL, l1[XS_WATCH_TOKEN]); XL_LOG(ctx, XL_LOG_DEBUG, "Destroyed device backend at %s", l1[XS_WATCH_TOKEN]); rc = 0; } - libxl_free(ctx, state); free(l...

This patch series introduces the tapback daemon, the user space daemon that acts as a device''s back-end, essentially most of blkback in user space. The daemon is responsible for coordinating the front-end and tapdisk. It creates tapdisk process as needed, instructs them to connect to/disconnect from the shared ring, and manages the state of the back-end. The shared ring between the

This patch series implements the Linux Xen guest in terms of the paravirt-ops interface. The features in implemented this patch series are: * domU only * UP only (most code is SMP-safe, but there's no way to create a new vcpu) * writable pagetables, with late pinning/early unpinning (no shadow pagetable support) * supports both PAE and non-PAE modes * xen console * virtual block

This patch series implements the Linux Xen guest in terms of the paravirt-ops interface. The features in implemented this patch series are: * domU only * UP only (most code is SMP-safe, but there's no way to create a new vcpu) * writable pagetables, with late pinning/early unpinning (no shadow pagetable support) * supports both PAE and non-PAE modes * xen console * virtual block

This patch series implements the Linux Xen guest in terms of the paravirt-ops interface. The features in implemented this patch series are: * domU only * UP only (most code is SMP-safe, but there's no way to create a new vcpu) * writable pagetables, with late pinning/early unpinning (no shadow pagetable support) * supports both PAE and non-PAE modes * xen console * virtual block

Hi Andi, This patch series implements the Linux Xen guest as a paravirt_ops backend. The features in implemented this patch series are: * domU only * UP only (most code is SMP-safe, but there's no way to create a new vcpu) * writable pagetables, with late pinning/early unpinning (no shadow pagetable support) * supports both PAE and non-PAE modes * xen hvc console (console=hvc0) *

Hi Andi, This patch series implements the Linux Xen guest as a paravirt_ops backend. The features in implemented this patch series are: * domU only * UP only (most code is SMP-safe, but there's no way to create a new vcpu) * writable pagetables, with late pinning/early unpinning (no shadow pagetable support) * supports both PAE and non-PAE modes * xen hvc console (console=hvc0) *