| Message ID | 20181106024239.7935-3-chaohong.guo@intel.com |
|---|---|
| State | Not Applicable |
| Delegated to: | David Miller |
| Headers | show |
| Series | [v3,1/4] Allow system to allocate IRQ 0 to ATA devices | expand |
Hi, On 11/06/2018 03:42 AM, Chaohong guo wrote: > Interrupt vector 0 is valid on some platform. falcon driver wrongly > treated interrupt vector 0 as invalid. We fix it in libata, and fix it > in this driver as well. > > Signed-off-by: Chaohong guo <chaohong.guo@intel.com> This patch should be folded into patch #1 (otherwise it breaks bisectability). There are also other libata drivers using polling mode that need updating: pata_palmld.c and pata_platform.c. > --- > drivers/ata/pata_falcon.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/ata/pata_falcon.c b/drivers/ata/pata_falcon.c > index 41e0d6a6cd05..a991053aee87 100644 > --- a/drivers/ata/pata_falcon.c > +++ b/drivers/ata/pata_falcon.c > @@ -171,7 +171,7 @@ static int pata_falcon_init_one(void) > (unsigned long)base + ATA_HD_CONTROL); > > /* activate */ > - return ata_host_activate(host, 0, NULL, 0, &pata_falcon_sht); > + return ata_host_activate(host, -1, NULL, 0, &pata_falcon_sht); > } > > module_init(pata_falcon_init_one); Best regards, -- Bartlomiej Zolnierkiewicz Samsung R&D Institute Poland Samsung Electronics
On 11/06/2018 01:53 PM, Bartlomiej Zolnierkiewicz wrote: >> Interrupt vector 0 is valid on some platform. falcon driver wrongly >> treated interrupt vector 0 as invalid. We fix it in libata, and fix it >> in this driver as well. >> >> Signed-off-by: Chaohong guo <chaohong.guo@intel.com> > > This patch should be folded into patch #1 (otherwise it breaks > bisectability). > > There are also other libata drivers using polling mode that need > updating: pata_palmld.c and pata_platform.c. Looks like there'd be a more subtle breaksage in sata_mv.c with the DT case -- we need to translate 0 returned from irq_of_parse_and_map() to a error code like -EINVAL... else we'd register IRQ0 when there's no IRQ. [...] > Best regards, > -- > Bartlomiej Zolnierkiewicz > Samsung R&D Institute Poland > Samsung Electronics MBR, Sergei
On 11/06/2018 05:42 AM, Chaohong guo wrote: > Interrupt vector 0 is valid on some platform. falcon driver wrongly > treated interrupt vector 0 as invalid. We fix it in libata, and fix it > in this driver as well. > > Signed-off-by: Chaohong guo <chaohong.guo@intel.com> > --- > drivers/ata/pata_falcon.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/ata/pata_falcon.c b/drivers/ata/pata_falcon.c > index 41e0d6a6cd05..a991053aee87 100644 > --- a/drivers/ata/pata_falcon.c > +++ b/drivers/ata/pata_falcon.c > @@ -171,7 +171,7 @@ static int pata_falcon_init_one(void) > (unsigned long)base + ATA_HD_CONTROL); > > /* activate */ > - return ata_host_activate(host, 0, NULL, 0, &pata_falcon_sht); > + return ata_host_activate(host, -1, NULL, 0, &pata_falcon_sht); I'd prefer smth like -EINVAL to -1. MBR, Sergei
On 11/06/2018 05:42 AM, Chaohong guo wrote: > Interrupt vector 0 is valid on some platform. falcon driver wrongly > treated interrupt vector 0 as invalid. We fix it in libata, and fix it > in this driver as well. > > Signed-off-by: Chaohong guo <chaohong.guo@intel.com> Oops, Bart and I missed this -- there should be no spaces before the patch tags. [...] MBR, Sergei
On 11/06/2018 06:04 PM, Sergei Shtylyov wrote: >>> Interrupt vector 0 is valid on some platform. falcon driver wrongly >>> treated interrupt vector 0 as invalid. We fix it in libata, and fix it >>> in this driver as well. >>> >>> Signed-off-by: Chaohong guo <chaohong.guo@intel.com> >> >> This patch should be folded into patch #1 (otherwise it breaks >> bisectability). >> >> There are also other libata drivers using polling mode that need >> updating: pata_palmld.c and pata_platform.c. > > Looks like there'd be a more subtle breaksage in sata_mv.c with the DT case -- > we need to translate 0 returned from irq_of_parse_and_map() to a error code like > -EINVAL... else we'd register IRQ0 when there's no IRQ. pata_samsung_cf.c also needs fixing in the same patch #1... > [...] >> Best regards, >> -- >> Bartlomiej Zolnierkiewicz >> Samsung R&D Institute Poland >> Samsung Electronics MBR, Sergei
On 11/07/2018 07:29 PM, Sergei Shtylyov wrote: >>>> Interrupt vector 0 is valid on some platform. falcon driver wrongly >>>> treated interrupt vector 0 as invalid. We fix it in libata, and fix it >>>> in this driver as well. >>>> >>>> Signed-off-by: Chaohong guo <chaohong.guo@intel.com> >>> >>> This patch should be folded into patch #1 (otherwise it breaks >>> bisectability). >>> >>> There are also other libata drivers using polling mode that need >>> updating: pata_palmld.c and pata_platform.c. >> >> Looks like there'd be a more subtle breaksage in sata_mv.c with the DT case -- >> we need to translate 0 returned from irq_of_parse_and_map() to a error code like >> -EINVAL... else we'd register IRQ0 when there's no IRQ. > > pata_samsung_cf.c also needs fixing in the same patch #1... And pata_octeon_cf.c too... >> [...] >>> Best regards, >>> -- >>> Bartlomiej Zolnierkiewicz >>> Samsung R&D Institute Poland >>> Samsung Electronics MBR, Sergei
diff --git a/drivers/ata/pata_falcon.c b/drivers/ata/pata_falcon.c index 41e0d6a6cd05..a991053aee87 100644 --- a/drivers/ata/pata_falcon.c +++ b/drivers/ata/pata_falcon.c @@ -171,7 +171,7 @@ static int pata_falcon_init_one(void) (unsigned long)base + ATA_HD_CONTROL); /* activate */ - return ata_host_activate(host, 0, NULL, 0, &pata_falcon_sht); + return ata_host_activate(host, -1, NULL, 0, &pata_falcon_sht); } module_init(pata_falcon_init_one);
Interrupt vector 0 is valid on some platform. falcon driver wrongly treated interrupt vector 0 as invalid. We fix it in libata, and fix it in this driver as well. Signed-off-by: Chaohong guo <chaohong.guo@intel.com> --- drivers/ata/pata_falcon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)