search for: free_connect_params

I believe this roughly implements Nir's proposal here: https://www.redhat.com/archives/libguestfs/2020-August/msg00028.html Unfortunately it doesn't work for me. It actually slows things down quite a lot, for reasons I don't understand. Note the adjustment of the pool-max parameter and how it affects the total time. The results are quite reproducible. $ ./nbdkit -r -U - vddk

...gt; +#define THREAD_MODEL NBDKIT_THREAD_MODEL_SERIALIZE_REQUESTS > + > +/* Lock protecting open/close calls - see above. */ > +static pthread_mutex_t open_close_lock = PTHREAD_MUTEX_INITIALIZER; > > /* The per-connection handle. */ > struct vddk_handle { > @@ -524,6 +531,7 @@ free_connect_params (VixDiskLibConnectParams *params) > static void * > vddk_open (int readonly) > { > + ACQUIRE_LOCK_FOR_CURRENT_SCOPE (&open_close_lock); > struct vddk_handle *h; > VixError err; > uint32_t flags; > @@ -616,6 +624,7 @@ vddk_open (int readonly) > static voi...

...DK disk handles. Do not access this directly, use + * GET_VDDK_HANDLE_FOR_CURRENT_SCOPE macro to get a free handle. + */ + pthread_mutex_t vddk_handles_lock; + pthread_cond_t vddk_handles_cond; + vddk_handles vddk_handles; }; static inline VixDiskLibConnectParams * @@ -531,17 +556,28 @@ free_connect_params (VixDiskLibConnectParams *params) static void * vddk_open (int readonly) { - ACQUIRE_LOCK_FOR_CURRENT_SCOPE (&open_close_lock); - struct vddk_handle *h; + struct handle *h; VixError err; - uint32_t flags; - h = malloc (sizeof *h); + h = calloc (1, sizeof *h); if (h == NULL) { -...

...ly, use > + * GET_VDDK_HANDLE_FOR_CURRENT_SCOPE macro to get a free handle. > + */ > + pthread_mutex_t vddk_handles_lock; > + pthread_cond_t vddk_handles_cond; > + vddk_handles vddk_handles; > }; > > static inline VixDiskLibConnectParams * > @@ -531,17 +556,28 @@ free_connect_params (VixDiskLibConnectParams *params) > static void * > vddk_open (int readonly) > { > - ACQUIRE_LOCK_FOR_CURRENT_SCOPE (&open_close_lock); > - struct vddk_handle *h; > + struct handle *h; > VixError err; > - uint32_t flags; > > - h = malloc (sizeof *h); &gt...

...-connection handle. */ struct vddk_handle { VixDiskLibConnectParams *params; /* connection parameters */ VixDiskLibConnection connection; /* connection */ VixDiskLibHandle handle; /* disk handle */ + int readonly; }; static inline VixDiskLibConnectParams * @@ -557,6 +569,7 @@ free_connect_params (VixDiskLibConnectParams *params) static void * vddk_open (int readonly) { + ACQUIRE_LOCK_FOR_CURRENT_SCOPE (&open_close_lock); struct vddk_handle *h; VixError err; uint32_t flags; @@ -567,6 +580,7 @@ vddk_open (int readonly) return NULL; } + h->readonly = readonly;...

..._THREAD_MODEL_SERIALIZE_ALL_REQUESTS +#define THREAD_MODEL NBDKIT_THREAD_MODEL_SERIALIZE_REQUESTS + +/* Lock protecting open/close calls - see above. */ +static pthread_mutex_t open_close_lock = PTHREAD_MUTEX_INITIALIZER; /* The per-connection handle. */ struct vddk_handle { @@ -524,6 +531,7 @@ free_connect_params (VixDiskLibConnectParams *params) static void * vddk_open (int readonly) { + ACQUIRE_LOCK_FOR_CURRENT_SCOPE (&open_close_lock); struct vddk_handle *h; VixError err; uint32_t flags; @@ -616,6 +624,7 @@ vddk_open (int readonly) static void vddk_close (void *handle) { + ACQUIRE_LO...

...uot;VixDiskLib_Open: %s", filename); + goto err2; + } + + nbdkit_debug ("transport mode for single_link: %s", + VixDiskLib_GetTransportMode (h->handle_single_link)); + } + return h; err2: @@ -559,6 +573,10 @@ vddk_close (void *handle) free_connect_params (h->params); DEBUG_CALL ("VixDiskLib_Close", "handle"); VixDiskLib_Close (h->handle); + if (single_link) { + DEBUG_CALL ("VixDiskLib_Close", "handle_single_link"); + VixDiskLib_Close (h->handle_single_link); + } DEBUG_CALL ("Vix...

In theory this patch depends on this series: https://www.redhat.com/archives/libguestfs/2020-August/msg00021.html In practice I believe they're independent of each other, but the above series makes it easier to test. Rich.

We had a query yesterday about the VDDK plugin and making it actually obey the weird "Multithreading Considerations" rules in the VDDK documentation (https://vdc-download.vmware.com/vmwb-repository/dcr-public/8f96698a-0e7b-4d67-bb6c-d18a1d101540/ef536a47-27cd-481a-90ef-76b38e75353c/vsphere-vddk-671-programming-guide.pdf) This patch is my attempt to implement this. The idea is that the

Couple of fixes to return correct data and one nice-to-have clean-up which is not needed. I just find it nicer to read. Martin Kletzander (3): vddk: Use a separate handle for single-link=true vddk: Do not report hole extents to be zero with single-link=true vddk: Eliminate one needless goto plugins/vddk/vddk.c | 48 +++++++++++++++++++++++++++++++++------------ 1 file changed, 36

This is my second attempt at this. The first version (also not working) was here: https://www.redhat.com/archives/libguestfs/2019-October/msg00062.html In part 1/2 I introduce a new .ready_to_serve plugin method which is called after forking and just before accepting any client connection. The idea would be that plugins could start background threads here. However this doesn't work well in