Skip to contentSkip to conversation

pty: Test data transmission with SLIP line discipline

PATCH Superseded Richard Palethorpe 2 participants 2 messages Feb 20, 2020 – Feb 25, 2020

v1 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 pty: Test data transmission with SLIP line discipline

Basic transmission test and try to trigger a possible race between receiving and hangup (it appears there is none from my testing).

This only includes SLIP for now, but more disciplines can be added. Probably at least CAN. However the packet generation will be different for each.

---

FYI, I considered patching the kernel so that SLIP sends unthrottle after each read, but on some physical serial lines this will cause one of the pins to change state which some crappy device was abusing for non-flow control purposes. Also it may interfer with flow control implemented somewhere else in the stack. I doubt this will really cause any problems, but OTOH I don't care about making SLIP work nicely with blocking PTYs enough to risk any change to the kernel. AFAICT there is no reason to use SLIP with a PTY except for virtual testing of SLIP.

 runtest/pty                     |   1 +
 testcases/kernel/pty/.gitignore |   1 +
 testcases/kernel/pty/pty04.c    | 250 ++++++++++++++++++++++++++++++++
 3 files changed, 252 insertions(+)
 create mode 100644 testcases/kernel/pty/pty04.c
runtest/pty +1 -0
index e484da0ff..5587312d3 100644
@@ -2,6 +2,7 @@
 pty01 pty01
 pty02 pty02
 pty03 pty03
+pty04 pty04
 ptem01 ptem01
 hangup01 hangup01
 
testcases/kernel/pty/.gitignore +1 -0
index 5b4f6a8c9..c67d723d2 100644
@@ -3,3 +3,4 @@
 /pty01
 /pty02
 /pty03
+/pty04
testcases/kernel/pty/pty04.c new file +250 -0

250 lines not on this page · raw message

Signature · 1 line
2.24.0
#1 lore raw

PATCH 1/1 pty: Test data transmission with SLIP line discipline

Quoted 8 lines · Richard Palethorpe

Basic transmission test and try to trigger a possible race between receiving and hangup (it appears there is none from my testing).

This only includes SLIP for now, but more disciplines can be added. Probably at least CAN. However the packet generation will be different for each.

---

...

testcases/kernel/pty/pty04.c

Quoted 3 lines
testcases/kernel/pty/pty04.c new file +0 -0
@@ -0,0 +1,250 @@

...> +struct ldisc_info {

testcases/kernel/pty/pty04.c line 56

Quoted 7 lines
testcases/kernel/pty/pty04.c new file +7 -0
+	int n;
+	char *name;
+	int max_mtu;
+};
+
+static struct ldisc_info ldiscs[] = {
+	{0, "TTY", 0},

Why not N_TTY?

Why do you define the line above btw?

testcases/kernel/pty/pty04.c line 57

Quoted 1 line
testcases/kernel/pty/pty04.c new file +1 -0
+	{1, "SLIP", 8192},

And N_SLIP?

Quoted 1 line
testcases/kernel/pty/pty04.c new file +1 -0
+};

...

Quoted 4 lines
testcases/kernel/pty/pty04.c new file +4 -0
+static void read_netdev(void)
+{
+	char *data = SAFE_MALLOC(mtu - 1);
+

testcases/kernel/pty/pty04.c line 191

Quoted 10 lines
testcases/kernel/pty/pty04.c new file +10 -0
+	tst_res(TINFO, "Reading from socket %d", sk);
+
+	SAFE_READ(1, sk, data, mtu - 1);
+	tst_res(TPASS, "Read netdev 1");
+	SAFE_READ(1, sk, data, mtu - 1);
+
+	tst_res(TPASS, "Read netdev 2");
+
+	TST_CHECKPOINT_WAKE(0);
+	while(read(sk, data, mtu - 1) > 0)

Could you check whether '_' is actually received here?

testcases/kernel/pty/pty04.c line 197

Quoted 6 lines
testcases/kernel/pty/pty04.c new file +6 -0
+		;
+
+	tst_res(TPASS, "Reading data from netdev interrupted by hangup");
+
+	free(data);
+}

thanks,

Signature · 2 lines
js
suse labs