[GSOC] System V IPC in userspace week3 report

[grigore larisa <larisagrigore@xxxxxxxxx>]

--047d7bd7657c5601ae04e0f401cc
Content-Type: text/plain; charset=ISO-8859-1

Hello,

This week I have done the following things:
- extended my tests in order to support multiple clients
- implemented a hashtable in order to find easy a client using its pid. It
can be used to verify if a client is already connected.
- resolved some bugs related to locking and polling
- investigated the impact of implementing shared memory in userland

These days I've studied how shm is implemented in kernel and how I could
move it in userland. Moving it in userland means moving only data
associated with each segment and permission checks. The operation of
allocating or mapping a segment must still be done in kernel.

As far as I understand, the project purpose is to implement in userland
those parts of sysv ipc resources that help big ipc consumers to have
better performance. I think this can be a good idea for semaphores and
message queues, where some syscalls can be avoided. In shared memory case,
I don't see that possible.
For shmget call, two messages must be sent (to the daemon and back to the
client) plus a syscall made by daemon when it must allocate a segment (when
some client need a shared memory resource that doesn't exist, such a
segment is allocated). For shmclt, one or two messages must be sent,
depending one the command. Maybe, only in shmat and shmdt I can avoid to
send messages to the daemon if some data are kept by the driver (number of
clients that use the resource for example) but the client will still do a
syscall to map/detach the shared memory object (as in the current
implementation).

In semaphores and message queues case, even if obtaining/controlling (*get,
*ctl) the object is more expensive than the kernel implementation because
of the communication with the daemon, semop()/msgget()/msgrcv() (that are
more frequently used) in userland are less expensive because they do
operations on shared memory.

I think is a better idea to implement userland semaphores and queues on top
of sysv shm already existing. What do you think about this?

Larisa

--047d7bd7657c5601ae04e0f401cc
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hello,<div><br></div><div style>This week I have done the =
following things:</div><div style>- extended my tests in order to support m=
ultiple clients</div><div style>- implemented a hashtable in order to find =
easy a client using its pid. It can be used to verify if a client is alread=
y connected.</div>
<div style>- resolved some bugs related to locking and polling</div><div st=
yle>- investigated the impact of implementing shared memory in userland</di=
v><div style><br></div><div style><div style=3D"font-family:arial,sans-seri=
f;font-size:13px">
These days I&#39;ve studied how shm is implemented in kernel and how I coul=
d move it in userland. Moving it in userland means moving only data associa=
ted with each segment and permission checks. The operation of allocating or=
 mapping a segment must still be done in kernel.</div>
<div style=3D"font-family:arial,sans-serif;font-size:13px"><br></div><div s=
tyle=3D"font-family:arial,sans-serif;font-size:13px">As far as I understand=
, the project purpose is to implement in userland those parts of sysv ipc r=
esources that help big ipc consumers to have better performance. I think th=
is can be a good idea for semaphores and message queues, where some syscall=
s can be avoided. In shared memory case, I don&#39;t see that possible.</di=
v>
<div style=3D"font-family:arial,sans-serif;font-size:13px">For shmget call,=
 two messages must be sent (to the daemon and back to the client) plus a sy=
scall made by daemon when it must allocate a segment (when some client need=
 a shared memory resource that doesn&#39;t exist, such a segment is allocat=
ed). For shmclt, one or two messages must be sent, depending one the comman=
d. Maybe, only in shmat and shmdt I can avoid to send messages to the daemo=
n if some data are kept by the driver (number of clients that use the resou=
rce for example) but the client will still do a syscall to map/detach the s=
hared memory object (as in the current implementation).</div>
<div style=3D"font-family:arial,sans-serif;font-size:13px"><br></div><div s=
tyle=3D"font-family:arial,sans-serif;font-size:13px">In semaphores and mess=
age queues case, even if obtaining/controlling (*get, *ctl) the object is m=
ore expensive than the kernel implementation because of the communication w=
ith the daemon, semop()/msgget()/msgrcv() (that are more frequently=A0used)=
 in userland are less expensive because they do operations on shared memory=
.</div>
<div style=3D"font-family:arial,sans-serif;font-size:13px"><br></div><div s=
tyle=3D"font-family:arial,sans-serif;font-size:13px">I think is a better id=
ea to implement userland semaphores and queues on top of sysv shm already e=
xisting. What do you think about this?</div>
<div style=3D"font-family:arial,sans-serif;font-size:13px"><br></div><div s=
tyle=3D"font-family:arial,sans-serif;font-size:13px">Larisa</div></div></di=
v>

--047d7bd7657c5601ae04e0f401cc--