If an application passes buffers to libiscsi for reading they are
currently added to a linked list one by one. This leads to a malloc
for each buffer object plus O(n) walks trough the list to add
the buffer to then end of the list. Additionally the buffer read
routine takes up to O(n) iterations to find the right buffer
for a request.
This patch introduces an scsi_iovector struct to pass buffers to
an scsi task. Adding a new buffer is in O(1) and finding the
right buffer to also. Malloc requirements are in O(log(n)).
Additionally the scsi_iovector struct is itended to be binary
compatible to an QEMUIOVector allowing to pass this structure
directly to the library.
Initial tests have been made booting an Ubuntu LTS 12.04.1
Desktop server up to the login prompt. The following observations
have been made with regards to scsi_malloc calls:
original implementation: ~11.500 mallocs
using iovector instead of list: ~ 7.500 mallocs
passing the iovector directly: 0 mallocs
To enable this feature in qemu for testing the following patch might
be used:
diff --git a/block/iscsi.c b/block/iscsi.c
index a6a819d..2809c15 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -390,11 +390,16 @@ iscsi_aio_readv(BlockDriverState *bs, int64_t sector_num,
return NULL;
}
+#if defined(LIBISCSI_FEATURE_IOVECTOR)
+ assert(sizeof(struct QEMUIOVector) == sizeof(struct scsi_iovector));
+ scsi_iovector_assign(acb->task, (struct scsi_iovector*) acb->qiov);
+#else
for (i = 0; i < acb->qiov->niov; i++) {
scsi_task_add_data_in_buffer(acb->task,
acb->qiov->iov[i].iov_len,
acb->qiov->iov[i].iov_base);
}
+#endif
iscsi_set_events(iscsilun);
---
Signed-off-by: Peter Lieven <pl@kamp.de>
Split the INQUIRY data-in unmarshalling into smaller chunks, fixing some
potential memory leaks along the way.
Signed-off-by: Arne Redlich <arne.redlich@googlemail.com>
The memory tracking code reports memory allocated by iscsi_allocate_pdu_with_itt_flags_size() as lost.
This memory is allocated by the iscsi part of libiscsi, but later freed by the lowlevel scsi part. We
will fix this later by introducing an iscsi_task object.
These two functions belong in the iscsi layer, not the scsi layer so move them
out from scsi-lowlevel.c so that we can start turning scsi-lowlevel.c to a pure
scsi layer and remove all dependencies to iscsi from it.
Change the list-head structure for in-task scsi memory allocations to
be private to scsi-lowlevel since is is never accessed from anyehwere else and
it is private to this function.
Remove the pointer to the user data in the list head and replace it with a zero length buffer at the end of the header.
Signed-off-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
In case there is an error condition e.g. out of memory. We are heavily
disconnecting and reconnecting without any limit. This patch
adds a 5 seconds period that has to go by between 2 reconnects.
DPRINTF was blindly sending all debug output to fd 2 (stderr).
The new function iscsi_set_debug_fd() will set the debug fd.
In general debugging is completely disabled by default.
If a process opens more than once connection the interfaces are assigned
round-robin from the available ones. However, different processes will
uses a random interface of as starting point. This patch will also
make the redirect case handled correctly.
This patch adds logarithmic malloc behaviour to iscsi_add_data().
Currently for each new call there is a new buffer allocated
and all old data is copied to the new buffer. Change this by
allocating at least PAGE_SIZE bytes and increase the allocation
by powers of 2 each time it does no longer fit.
This patch adds a wrapper around all memory allocations and frees.
The idea is to get warned immediately if the application leaks memory.
Additionally the wrapper functions make it easy to add different
memory allocators or memory pools in the future.
