# \[PATCH\] ata: hpt366: fix incorrect mask when checking at cmd\_high\_time - Subsystems: LIBATA SUBSYSTEM (Serial and Parallel ATA drivers), THE REST - Maintainers: Damien Le Moal, Niklas Cassel, Linus Torvalds - List: linux-ide - 5 messages, 4 authors - Page: https://ratatoskr.run/linux-ide/2016/07/4792802/t ## \[PATCH\] ata: hpt366: fix incorrect mask when checking at cmd\_high\_time - From: Colin King \ - Date: 2016-07-12 11:17:06 - Message: https://ratatoskr.run/linux-ide/2016/07/4792802.md ```text From: Colin Ian King According to the HPT366 data sheet, PCI config space dword 0x40-0x43 bits 11:8 specify the primary drive cmd_high_time, however, currently just 3 bits of the 4 are being used because the mask is 0x700 and not 0x0f00. Fix the mask, allowing for the 40MHz clock to be detected. Signed-off-by: Colin Ian King --- drivers/ata/pata_hpt366.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c index e5fb752..a219a50 100644 --- a/drivers/ata/pata_hpt366.c +++ b/drivers/ata/pata_hpt366.c @@ -368,7 +368,7 @@ static int hpt36x_init_one(struct pci_dev *dev, const struct pci_device_id *id) /* PCI clocking determines the ATA timing values to use */ /* info_hpt366 is safe against re-entry so we can scribble on it */ - switch ((reg1 & 0x700) >> 8) { + switch ((reg1 & 0xf00) >> 8) { case 9: hpriv = &hpt366_40; break; -- 2.8.1 ``` ## Re: \[PATCH\] ata: hpt366: fix incorrect mask when checking at cmd\_high\_time - From: Bartlomiej Zolnierkiewicz \ - Date: 2016-07-12 11:28:34 - Message: https://ratatoskr.run/linux-ide/2016/07/4792809.md ```text Hi, On Tuesday, July 12, 2016 12:16:19 PM Colin King wrote: > From: Colin Ian King > > According to the HPT366 data sheet, PCI config space dword 0x40-0x43 > bits 11:8 specify the primary drive cmd_high_time, however, > currently just 3 bits of the 4 are being used because the mask > is 0x700 and not 0x0f00. Fix the mask, allowing for the 40MHz clock > to be detected. > > Signed-off-by: Colin Ian King Acked-by: Bartlomiej Zolnierkiewicz Thanks for the patch, could you also fix also the old driver (drivers/ide/hpt366.c)? Best regards, -- Bartlomiej Zolnierkiewicz Samsung R&D Institute Poland Samsung Electronics > --- > drivers/ata/pata_hpt366.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c > index e5fb752..a219a50 100644 > --- a/drivers/ata/pata_hpt366.c > +++ b/drivers/ata/pata_hpt366.c > @@ -368,7 +368,7 @@ static int hpt36x_init_one(struct pci_dev *dev, const struct pci_device_id *id) > > /* PCI clocking determines the ATA timing values to use */ > /* info_hpt366 is safe against re-entry so we can scribble on it */ > - switch ((reg1 & 0x700) >> 8) { > + switch ((reg1 & 0xf00) >> 8) { > case 9: > hpriv = &hpt366_40; > break; ``` ## Re: \[PATCH\] ata: hpt366: fix incorrect mask when checking at cmd\_high\_time - From: Colin Ian King \ - Date: 2016-07-12 11:33:53 - Message: https://ratatoskr.run/linux-ide/2016/07/4792812.md ```text On 12/07/16 12:27, Bartlomiej Zolnierkiewicz wrote: > > Hi, > > On Tuesday, July 12, 2016 12:16:19 PM Colin King wrote: >> From: Colin Ian King >> >> According to the HPT366 data sheet, PCI config space dword 0x40-0x43 >> bits 11:8 specify the primary drive cmd_high_time, however, >> currently just 3 bits of the 4 are being used because the mask >> is 0x700 and not 0x0f00. Fix the mask, allowing for the 40MHz clock >> to be detected. >> >> Signed-off-by: Colin Ian King > > Acked-by: Bartlomiej Zolnierkiewicz > > Thanks for the patch, could you also fix also the old driver > (drivers/ide/hpt366.c)? Yep, patch already sent. https://lkml.org/lkml/2016/7/12/199 > > Best regards, > -- > Bartlomiej Zolnierkiewicz > Samsung R&D Institute Poland > Samsung Electronics > >> --- >> drivers/ata/pata_hpt366.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c >> index e5fb752..a219a50 100644 >> --- a/drivers/ata/pata_hpt366.c >> +++ b/drivers/ata/pata_hpt366.c >> @@ -368,7 +368,7 @@ static int hpt36x_init_one(struct pci_dev *dev, const struct pci_device_id *id) >> >> /* PCI clocking determines the ATA timing values to use */ >> /* info_hpt366 is safe against re-entry so we can scribble on it */ >> - switch ((reg1 & 0x700) >> 8) { >> + switch ((reg1 & 0xf00) >> 8) { >> case 9: >> hpriv = &hpt366_40; >> break; > ``` ## Re: \[PATCH\] ata: hpt366: fix incorrect mask when checking at cmd\_high\_time - From: Tejun Heo \ - Date: 2016-07-12 15:03:37 - Message: https://ratatoskr.run/linux-ide/2016/07/4793014.md ```text On Tue, Jul 12, 2016 at 12:16:19PM +0100, Colin King wrote: > From: Colin Ian King > > According to the HPT366 data sheet, PCI config space dword 0x40-0x43 > bits 11:8 specify the primary drive cmd_high_time, however, > currently just 3 bits of the 4 are being used because the mask > is 0x700 and not 0x0f00. Fix the mask, allowing for the 40MHz clock > to be detected. > > Signed-off-by: Colin Ian King Applied to libata/for-4.8. Thanks. -- tejun ``` ## Re: \[PATCH\] ata: hpt366: fix incorrect mask when checking at cmd\_high\_time - From: One Thousand Gnomes \ - Date: 2016-07-18 20:15:02 - Message: https://ratatoskr.run/linux-ide/2016/07/4797379.md ```text On Tue, 12 Jul 2016 12:16:19 +0100 Colin King [ref] wrote: > From: Colin Ian King > > According to the HPT366 data sheet, PCI config space dword 0x40-0x43 > bits 11:8 specify the primary drive cmd_high_time, however, > currently just 3 bits of the 4 are being used because the mask > is 0x700 and not 0x0f00. Fix the mask, allowing for the 40MHz clock > to be detected. Is this tested on real hardware ? I learned long ago never to blindly trust IDE data sheets. It looks right but if anyone actually has a 40MHz PCI bus box with an HPT366 it would be good to test. Alan ```