Thread (11 messages) 11 messages, 5 authors, 2010-02-02

Re: [PATCH 2/4] mpc8569mds: Add bscr setting for rtbi mode

From: Grant Likely <hidden>
Date: 2010-02-01 21:57:53
Also in: netdev

On Thu, Jan 14, 2010 at 1:13 AM, Liu Yu [off-list ref] wrote:
Signed-off-by: Liu Yu <redacted>
---
=A0arch/powerpc/platforms/85xx/mpc85xx_mds.c | =A0 24 +++++++++++++++++++=
+++++
quoted hunk ↗ jump to hunk
=A01 files changed, 24 insertions(+), 0 deletions(-)
diff --git a/arch/powerpc/platforms/85xx/mpc85xx_mds.c b/arch/powerpc/pla=
tforms/85xx/mpc85xx_mds.c
quoted hunk ↗ jump to hunk
index c5028a2..0872e4a 100644
--- a/arch/powerpc/platforms/85xx/mpc85xx_mds.c
+++ b/arch/powerpc/platforms/85xx/mpc85xx_mds.c
@@ -237,6 +237,8 @@ static void __init mpc85xx_mds_setup_arch(void)
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0} else if (machine_is(mpc8569_mds)) {
=A0#define BCSR7_UCC12_GETHnRST =A0 (0x1 << 2)
=A0#define BCSR8_UEM_MARVELL_RST =A0(0x1 << 1)
+#define BCSR_UCC_RGMII =A0 =A0 =A0 =A0 (0x1 << 6)
+#define BCSR_UCC_RTBI =A0 =A0 =A0 =A0 =A0(0x1 << 5)
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0/*
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 * U-Boot mangles interrup=
t polarity for Marvell PHYs,
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 * so reset built-in and U=
EM Marvell PHYs, this puts
quoted hunk ↗ jump to hunk
@@ -247,6 +249,28 @@ static void __init mpc85xx_mds_setup_arch(void)
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0setbits8(&bcsr_regs[7], BC=
SR7_UCC12_GETHnRST);
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0clrbits8(&bcsr_regs[8], BC=
SR8_UEM_MARVELL_RST);
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 for (np =3D NULL; (np =3D o=
f_find_compatible_node(np,
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 "network",

Don't match on the 'type' field.  Replace "network" with NULL and just
rely on "ucc_geth" for matching.
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 "ucc_geth")) !=3D NULL;) {
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 const unsig=
ned int *prop;

u32 please.

Also, rather than reusing 'prop' for both char* and u32 values, which
forces you to use ugly casts, use 2 local variables here.
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 int ucc_num=
;
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 prop =3D of=
_get_property(np, "cell-index", NULL);
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (prop =
=3D=3D NULL)
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
 =A0 continue;
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 ucc_num =3D=
 *prop - 1;

Ugh.  No bounds checking...
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 prop =3D of=
_get_property(np, "phy-connection-type", NULL);
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (prop =
=3D=3D NULL)
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
 =A0 continue;
+
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 if (strcmp(=
"rtbi", (const char *)prop) =3D=3D 0)

(This is the ugly cast I was talking about.)
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=
 =A0 clrsetbits_8(&bcsr_regs[7 + ucc_num],

...not having bounds checking could result in badness in this array index.

This patch is dangerous as written.

Finally, while using cell-index seems convenient, I think it would be
better to have a lookup table of the index into the BCSR register
block from the UCC base address, which also gives you implicit bounds
checking.

g.

--=20
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help