| Message ID | 1428579988-10167-2-git-send-email-lambert.quentin@gmail.com |
|---|---|
| State | Superseded |
| Delegated to: | David Miller |
| Headers | show |
Why are these GFP_ATOMIC allocations? regards, dan carpenter -- To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Thu, 9 Apr 2015 13:46:27 +0200 Quentin Lambert <lambert.quentin@gmail.com> wrote: > Replace occurences of the pci api by appropriate call to the dma api. > Drivers/ide is obsolete. It probably ought to go at this point but even so it exists solely in case of a problem or compatibility issue with drivers/ata. Cleaning it up without testing all the drivers makes no sense IMHO. Cleaning it up and testing them doesn't make much sense either. Alan -- To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
From: One Thousand Gnomes <gnomes@lxorguk.ukuu.org.uk> Date: Mon, 13 Apr 2015 13:25:30 +0100 > On Thu, 9 Apr 2015 13:46:27 +0200 > Quentin Lambert <lambert.quentin@gmail.com> wrote: > >> Replace occurences of the pci api by appropriate call to the dma api. >> > > Drivers/ide is obsolete. It probably ought to go at this point but even > so it exists solely in case of a problem or compatibility issue with > drivers/ata. Cleaning it up without testing all the drivers makes no > sense IMHO. Cleaning it up and testing them doesn't make much sense > either. It's a straightforward transformation, that actually evaluates to what the definitions of the deprecated pci DMA interfaces are defined to. So I intend to apply this patch, especially for the sake of allowing these PCI DMA interfaces to be removed. -- To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Mon, Apr 13, 2015 at 01:08:13PM -0400, David Miller wrote: > From: One Thousand Gnomes <gnomes@lxorguk.ukuu.org.uk> > Date: Mon, 13 Apr 2015 13:25:30 +0100 > > > On Thu, 9 Apr 2015 13:46:27 +0200 > > Quentin Lambert <lambert.quentin@gmail.com> wrote: > > > >> Replace occurences of the pci api by appropriate call to the dma api. > >> > > > > Drivers/ide is obsolete. It probably ought to go at this point but even > > so it exists solely in case of a problem or compatibility issue with > > drivers/ata. Cleaning it up without testing all the drivers makes no > > sense IMHO. Cleaning it up and testing them doesn't make much sense > > either. > > It's a straightforward transformation, that actually evaluates to what > the definitions of the deprecated pci DMA interfaces are defined to. Almost, except that it uses GFP_KERNEL instead of GFP_DMA. But I reviewed it and I think it's fine. Quentin sent a GFP_ATOMIC patch as the first version and I asked him to redo it. Maybe I shouldn't have done that... regards, dan carpenter -- To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
From: Dan Carpenter <dan.carpenter@oracle.com> Date: Mon, 13 Apr 2015 20:23:24 +0300 > On Mon, Apr 13, 2015 at 01:08:13PM -0400, David Miller wrote: >> From: One Thousand Gnomes <gnomes@lxorguk.ukuu.org.uk> >> Date: Mon, 13 Apr 2015 13:25:30 +0100 >> >> > On Thu, 9 Apr 2015 13:46:27 +0200 >> > Quentin Lambert <lambert.quentin@gmail.com> wrote: >> > >> >> Replace occurences of the pci api by appropriate call to the dma api. >> >> >> > >> > Drivers/ide is obsolete. It probably ought to go at this point but even >> > so it exists solely in case of a problem or compatibility issue with >> > drivers/ata. Cleaning it up without testing all the drivers makes no >> > sense IMHO. Cleaning it up and testing them doesn't make much sense >> > either. >> >> It's a straightforward transformation, that actually evaluates to what >> the definitions of the deprecated pci DMA interfaces are defined to. > > Almost, except that it uses GFP_KERNEL instead of GFP_DMA. But I > reviewed it and I think it's fine. Quentin sent a GFP_ATOMIC patch as > the first version and I asked him to redo it. Maybe I shouldn't have > done that... No, I'm fine with that part too, don't worry. -- To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
--- a/drivers/ide/cs5520.c +++ b/drivers/ide/cs5520.c @@ -123,7 +123,7 @@ static int cs5520_init_one(struct pci_de return -ENODEV; } pci_set_master(dev); - if (pci_set_dma_mask(dev, DMA_BIT_MASK(32))) { + if (dma_set_mask(&dev->dev, DMA_BIT_MASK(32))) { printk(KERN_WARNING "%s: No suitable DMA available.\n", d->name); return -ENODEV; --- a/drivers/ide/pmac.c +++ b/drivers/ide/pmac.c @@ -1689,10 +1689,9 @@ static int pmac_ide_init_dma(ide_hwif_t * The +2 is +1 for the stop command and +1 to allow for * aligning the start address to a multiple of 16 bytes. */ - pmif->dma_table_cpu = pci_alloc_consistent( - dev, + pmif->dma_table_cpu = dma_alloc_coherent(&dev->dev, (MAX_DCMDS + 2) * sizeof(struct dbdma_cmd), - &hwif->dmatable_dma); + &hwif->dmatable_dma, GFP_ATOMIC); if (pmif->dma_table_cpu == NULL) { printk(KERN_ERR "%s: unable to allocate DMA command list\n", hwif->name); --- a/drivers/ide/setup-pci.c +++ b/drivers/ide/setup-pci.c @@ -209,7 +209,7 @@ static int ide_pci_enable(struct pci_dev * a DMA mask field to the struct ide_port_info if we need it * (or let lower level driver set the DMA mask) */ - ret = pci_set_dma_mask(dev, DMA_BIT_MASK(32)); + ret = dma_set_mask(&dev->dev, DMA_BIT_MASK(32)); if (ret < 0) { printk(KERN_ERR "%s %s: can't set DMA mask\n", d->name, pci_name(dev)); --- a/drivers/ide/sgiioc4.c +++ b/drivers/ide/sgiioc4.c @@ -334,8 +334,8 @@ static int ide_dma_sgiioc4(ide_hwif_t *h if (ide_allocate_dma_engine(hwif)) goto dma_pci_alloc_failure; - pad = pci_alloc_consistent(dev, IOC4_IDE_CACHELINE_SIZE, - (dma_addr_t *)&hwif->extra_base); + pad = dma_alloc_coherent(&dev->dev, IOC4_IDE_CACHELINE_SIZE, + (dma_addr_t *)&hwif->extra_base, GFP_ATOMIC); if (pad) { ide_set_hwifdata(hwif, pad); return 0;
Replace occurences of the pci api by appropriate call to the dma api. A simplified version of the semantic patch that finds this problem is as follows: (http://coccinelle.lip6.fr) @deprecated@ idexpression id; position p; @@ ( pci_dma_supported@p ( id, ...) | pci_alloc_consistent@p ( id, ...) ) @bad1@ idexpression id; position deprecated.p; @@ ...when != &id->dev when != pci_get_drvdata ( id ) when != pci_enable_device ( id ) ( pci_dma_supported@p ( id, ...) | pci_alloc_consistent@p ( id, ...) ) @depends on !bad1@ idexpression id; expression direction; position deprecated.p; @@ ( - pci_dma_supported@p ( id, + dma_supported ( &id->dev, ... + , GFP_ATOMIC ) | - pci_alloc_consistent@p ( id, + dma_alloc_coherent ( &id->dev, ... + , GFP_ATOMIC ) ) Signed-off-by: Quentin Lambert <lambert.quentin@gmail.com> --- drivers/ide/cs5520.c | 2 +- drivers/ide/pmac.c | 5 ++--- drivers/ide/setup-pci.c | 2 +- drivers/ide/sgiioc4.c | 4 ++-- 4 files changed, 6 insertions(+), 7 deletions(-) -- To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html