drivers/net/ethernet/xscale/ptp_ixp46x.c:283 ixp46x_ptp_find() warn: variable dereferenced before check 'ixp_clock.ptp_clock' (see line 281)
PATCH 17/23 drivers/net/ethernet/xscale/ptp_ixp46x.c:283 ixp46x_ptp_find() warn: variable dereferenced before check 'ixp_clock.ptp_clock' (see line 281)
Hi Arnd,
First bad commit (maybe != root cause):
tree: https://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-integrator.git ixp4xx-test head: 1b826e686b390fd1eeb347def38185fefcbd5140 commit: 7ac6c27d7ee1cdef95314be43e4c3854d8f80240 [17/23] ixp4xx_eth: enable compile testing
If you fix the issue, kindly add following tag
smatch warnings:drivers/net/ethernet/xscale/ptp_ixp46x.c:283 ixp46x_ptp_find() warn: variable dereferenced before check 'ixp_clock.ptp_clock' (see line 281)
# https://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-integrator.git/commit/?id=7ac6c27d7ee1cdef95314be43e4c3854d8f80240
git remote add arm-integrator https://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-integrator.git
git remote update arm-integrator
git checkout 7ac6c27d7ee1cdef95314be43e4c3854d8f80240
vim +283 drivers/net/ethernet/xscale/ptp_ixp46x.c
768672ff6ea477 drivers/net/ethernet/xscale/ptp_ixp46x.c Arnd Bergmann 2019-08-26 278 int ixp46x_ptp_find(struct ixp46x_ts_regs *__iomem *regs, int *phc_index)
32bd93e8f924c2 drivers/ptp/ptp_ixp46x.c Richard Cochran 2011-04-22 279 {
768672ff6ea477 drivers/net/ethernet/xscale/ptp_ixp46x.c Arnd Bergmann 2019-08-26 280 *regs = ixp_clock.regs;
768672ff6ea477 drivers/net/ethernet/xscale/ptp_ixp46x.c Arnd Bergmann 2019-08-26 @281 *phc_index = ptp_clock_index(ixp_clock.ptp_clock);
^^^^^^^^^^^^^^^^^^^Dereferenced inside function
768672ff6ea477 drivers/net/ethernet/xscale/ptp_ixp46x.c Arnd Bergmann 2019-08-26 282768672ff6ea477 drivers/net/ethernet/xscale/ptp_ixp46x.c Arnd Bergmann 2019-08-26 @283 if (!ixp_clock.ptp_clock)
^^^^^^^^^^^^^^^^^^^^
Too late.
768672ff6ea477 drivers/net/ethernet/xscale/ptp_ixp46x.c Arnd Bergmann 2019-08-26 284 return -EPROBE_DEFER; 768672ff6ea477 drivers/net/ethernet/xscale/ptp_ixp46x.c Arnd Bergmann 2019-08-26 285 768672ff6ea477 drivers/net/ethernet/xscale/ptp_ixp46x.c Arnd Bergmann 2019-08-26 286 return 0; 32bd93e8f924c2 drivers/ptp/ptp_ixp46x.c Richard Cochran 2011-04-22 287 } 768672ff6ea477 drivers/net/ethernet/xscale/ptp_ixp46x.c Arnd Bergmann 2019-08-26 288 EXPORT_SYMBOL_GPL(ixp46x_ptp_find);
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
Quoted 9 lines · Dan Carpenter
Hi Arnd,
First bad commit (maybe != root cause):
tree: https://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-integrator.git ixp4xx-test head: 1b826e686b390fd1eeb347def38185fefcbd5140 commit: 7ac6c27d7ee1cdef95314be43e4c3854d8f80240 [17/23] ixp4xx_eth: enable compile testing
Quoted 7 lines
If you fix the issue, kindly add following tag
Reported-by kbuild test robot <[email protected]>Reported-by Dan Carpenter <[email protected]>smatch warnings:
drivers/net/ethernet/xscale/ptp_ixp46x.c:283ixp46x_ptp_find()warn: variable dereferenced before check 'ixp_clock.ptp_clock' (see line 281)
This is clearly a bug I introduced, thanks a lot for the report!
Again, there is a trivial fix that Linus or I should fold into the original patch:
(quoted diff) +0 -0
b/drivers/net/ethernet/xscale/ptp_ixp46x.c
index bb62c265bb4d..2af8e47dc13b 100644
drivers/net/ethernet/xscale/ptp_ixp46x.c +3 -3
@@ -277,12 +277,12 @@ static int setup_interrupt(int gpio)int ixp46x_ptp_find (struct ixp46x_ts_regs *__iomem *regs,int *phc_index) { - *regs = ixp_clock.regs; - *phc_index =ptp_clock_index (ixp_clock.ptp_clock); -if (!ixp_clock.ptp_clock)return -EPROBE_DEFER ; + *regs = ixp_clock.regs; + *phc_index =ptp_clock_index (ixp_clock.ptp_clock); +return 0 ; }EXPORT_SYMBOL_GPL (ixp46x_ptp_find);
Arnd
Quoted 8 lines · Arnd Bergmann
Quoted 3 lines
smatch warnings:
drivers/net/ethernet/xscale/ptp_ixp46x.c:283ixp46x_ptp_find()warn: variable dereferenced before check 'ixp_clock.ptp_clock' (see line 281)This is clearly a bug I introduced, thanks a lot for the report!
Again, there is a trivial fix that Linus or I should fold into the original patch:
I folded it into the patch in my tree.
I'm working to get some of this in a shape that I can start funneling it to ARM-SoC.
Yours,
Linus Walleij