From: Ben Widawsky <ben.widawsky@intel.com>
To: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Cc: linux-cxl@vger.kernel.org, linux-pci@vger.kernel.org,
linux-acpi@vger.kernel.org, ira.weiny@intel.com,
vishal.l.verma@intel.com, alison.schofield@intel.com,
dan.j.williams@intel.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 4/7] cxl/mem: Get rid of @cxlm.base
Date: Tue, 13 Apr 2021 09:17:26 -0700 [thread overview]
Message-ID: <20210413161726.tz7rg46krrekk3lp@intel.com> (raw)
In-Reply-To: <20210408182635.00003997@Huawei.com>
On 21-04-08 18:26:35, Jonathan Cameron wrote:
> On Wed, 7 Apr 2021 15:26:22 -0700
> Ben Widawsky <ben.widawsky@intel.com> wrote:
>
> > @cxlm.base only existed to support holding the base found in the
> > register block mapping code, and pass it along to the register setup
> > code. Now that the register setup function has all logic around managing
> > the registers, from DVSEC to iomapping up to populating our CXL specific
> > information, it is easy to turn the @base values into local variables
> > and remove them from our device driver state.
> >
> > Signed-off-by: Ben Widawsky <ben.widawsky@intel.com>
>
> Patch is basically fine, but I do wonder if you could avoid the
> nasty casting in and out of __iomem in the error paths.
>
> It's a common enough idiom though so I'm not htat fussed.
>
> Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
>
> > ---
> > drivers/cxl/mem.c | 24 +++++++++++-------------
> > drivers/cxl/mem.h | 2 --
> > 2 files changed, 11 insertions(+), 15 deletions(-)
> >
> > diff --git a/drivers/cxl/mem.c b/drivers/cxl/mem.c
> > index 04b4f7445083..60b95c524c3e 100644
> > --- a/drivers/cxl/mem.c
> > +++ b/drivers/cxl/mem.c
> > @@ -922,11 +922,10 @@ static struct cxl_mem *cxl_mem_create(struct pci_dev *pdev)
> > return cxlm;
> > }
> >
> > -static int cxl_mem_map_regblock(struct cxl_mem *cxlm, u32 reg_lo, u32 reg_hi)
> > +static void __iomem *cxl_mem_map_regblock(struct cxl_mem *cxlm, u32 reg_lo, u32 reg_hi)
> > {
> > struct pci_dev *pdev = cxlm->pdev;
> > struct device *dev = &pdev->dev;
> > - void __iomem *regs;
> > u64 offset;
> > u8 bar;
> > int rc;
> > @@ -938,20 +937,18 @@ static int cxl_mem_map_regblock(struct cxl_mem *cxlm, u32 reg_lo, u32 reg_hi)
> > if (pci_resource_len(pdev, bar) < offset) {
> > dev_err(dev, "BAR%d: %pr: too small (offset: %#llx)\n", bar,
> > &pdev->resource[bar], (unsigned long long)offset);
> > - return -ENXIO;
> > + return (void __iomem *)ERR_PTR(-ENXIO);
> > }
> >
> > rc = pcim_iomap_regions(pdev, BIT(bar), pci_name(pdev));
> > if (rc) {
> > dev_err(dev, "failed to map registers\n");
> > - return rc;
> > + return (void __iomem *)ERR_PTR(rc);
>
> The casting is fairly horrible, perhaps just pass in
> a void __iomem ** and pass base back through that?
>
TIL: IOMEM_ERR_PTR. Would that suffice?
> > }
> > - regs = pcim_iomap_table(pdev)[bar];
> > -
> > - cxlm->base = regs + offset;
> >
> > dev_dbg(dev, "Mapped CXL Memory Device resource\n");
> > - return 0;
> > +
> > + return pcim_iomap_table(pdev)[bar] + offset;
> > }
> >
> > static int cxl_mem_dvsec(struct pci_dev *pdev, int dvsec)
> > @@ -993,7 +990,8 @@ static int cxl_mem_setup_regs(struct cxl_mem *cxlm)
> > struct pci_dev *pdev = cxlm->pdev;
> > struct device *dev = &pdev->dev;
> > u32 regloc_size, regblocks;
> > - int rc, regloc, i;
> > + void __iomem *base;
> > + int regloc, i;
> >
> > regloc = cxl_mem_dvsec(pdev, PCI_DVSEC_ID_CXL_REGLOC_OFFSET);
> > if (!regloc) {
> > @@ -1019,9 +1017,9 @@ static int cxl_mem_setup_regs(struct cxl_mem *cxlm)
> > reg_type = FIELD_GET(CXL_REGLOC_RBI_MASK, reg_lo);
> >
> > if (reg_type == CXL_REGLOC_RBI_MEMDEV) {
> > - rc = cxl_mem_map_regblock(cxlm, reg_lo, reg_hi);
> > - if (rc)
> > - return rc;
> > + base = cxl_mem_map_regblock(cxlm, reg_lo, reg_hi);
> > + if (IS_ERR(base))
> > + return PTR_ERR(base);
> > break;
> > }
> > }
> > @@ -1031,7 +1029,7 @@ static int cxl_mem_setup_regs(struct cxl_mem *cxlm)
> > return -ENXIO;
> > }
> >
> > - cxl_setup_device_regs(dev, cxlm->base, ®s->device_regs);
> > + cxl_setup_device_regs(dev, base, ®s->device_regs);
> >
> > if (!regs->status || !regs->mbox || !regs->memdev) {
> > dev_err(dev, "registers not found: %s%s%s\n",
> > diff --git a/drivers/cxl/mem.h b/drivers/cxl/mem.h
> > index 8bad7166adba..bfcfef461b16 100644
> > --- a/drivers/cxl/mem.h
> > +++ b/drivers/cxl/mem.h
> > @@ -49,7 +49,6 @@ struct cxl_memdev {
> > /**
> > * struct cxl_mem - A CXL memory device
> > * @pdev: The PCI device associated with this CXL device.
> > - * @base: IO mappings to the device's MMIO
> > * @cxlmd: Logical memory device chardev / interface
> > * @regs: Parsed register blocks
> > * @payload_size: Size of space for payload
> > @@ -62,7 +61,6 @@ struct cxl_memdev {
> > */
> > struct cxl_mem {
> > struct pci_dev *pdev;
> > - void __iomem *base;
> > struct cxl_memdev *cxlmd;
> >
> > struct cxl_regs regs;
>
next prev parent reply other threads:[~2021-04-13 16:25 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-07 22:26 [PATCH 0/7] Enumerate HDM Decoder registers Ben Widawsky
2021-04-07 22:26 ` [PATCH 1/7] cxl/mem: Use dev instead of pdev->dev Ben Widawsky
2021-04-08 17:08 ` Jonathan Cameron
2021-04-07 22:26 ` [PATCH 2/7] cxl/mem: Split creation from mapping in probe Ben Widawsky
2021-04-08 17:13 ` Jonathan Cameron
2021-04-07 22:26 ` [PATCH 3/7] cxl/mem: Move register locator logic into reg setup Ben Widawsky
2021-04-08 17:27 ` Jonathan Cameron
2021-04-07 22:26 ` [PATCH 4/7] cxl/mem: Get rid of @cxlm.base Ben Widawsky
2021-04-08 17:26 ` Jonathan Cameron
2021-04-13 16:17 ` Ben Widawsky [this message]
2021-04-14 9:24 ` Jonathan Cameron
2021-05-20 21:29 ` [PATCH v2 " Ben Widawsky
2021-04-07 22:26 ` [PATCH 5/7] cxl/mem: Move device register setup Ben Widawsky
2021-04-08 17:28 ` Jonathan Cameron
2021-04-07 22:26 ` [PATCH 6/7] cxl/mem: Create a helper to setup device regs Ben Widawsky
2021-04-08 17:33 ` Jonathan Cameron
2021-04-15 22:54 ` Dan Williams
2021-04-07 22:26 ` [PATCH 7/7] cxl: Add HDM decoder capbilities Ben Widawsky
2021-04-08 17:57 ` Jonathan Cameron
2021-04-15 23:27 ` Dan Williams
2021-04-15 23:50 ` Ben Widawsky
2021-04-16 0:25 ` Dan Williams
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210413161726.tz7rg46krrekk3lp@intel.com \
--to=ben.widawsky@intel.com \
--cc=Jonathan.Cameron@huawei.com \
--cc=alison.schofield@intel.com \
--cc=dan.j.williams@intel.com \
--cc=ira.weiny@intel.com \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-cxl@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=vishal.l.verma@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).