Linux CXL
 help / color / mirror / Atom feed
From: Dave Jiang <dave.jiang@intel.com>
To: Alison Schofield <alison.schofield@intel.com>,
	Davidlohr Bueso <dave@stgolabs.net>,
	Jonathan Cameron <jonathan.cameron@huawei.com>,
	Vishal Verma <vishal.l.verma@intel.com>,
	Ira Weiny <ira.weiny@intel.com>,
	Dan Williams <dan.j.williams@intel.com>
Cc: linux-cxl@vger.kernel.org
Subject: Re: [PATCH v2 1/3] cxl/region: Refactor address translation funcs for testing
Date: Thu, 4 Sep 2025 15:05:00 -0700	[thread overview]
Message-ID: <6deca3ef-2284-433f-ae91-ce1ff9f7a8b7@intel.com> (raw)
In-Reply-To: <f0c6fa733cea8585f2bfda0851489bd7ca2e2b2c.1756446925.git.alison.schofield@intel.com>



On 8/29/25 12:21 AM, Alison Schofield wrote:
> In preparation for adding a test module that exercises the address
> translation calculations, extract the core calculations into stand-
> alone functions that operate on base parameters without dependencies
> on struct cxl_region.
> 
> Mark the new functions as static outside of test builds by adding
> and using a new __mock_export label.
> 
> This refactoring enables unit testing of the address translation logic
> with controlled inputs, while maintaining identical functionality in
> the existing code paths.
> 
> The moved code has only one change. In the new cxl_calculate_position()
> eiw_to_ways(eiw, &ways) replaces the prior usage of p->interleave_ways,
> since the new function cannot depend upon struct cxl_region_params.
> 
> Signed-off-by: Alison Schofield <alison.schofield@intel.com>
> ---
>  drivers/cxl/core/region.c | 147 ++++++++++++++++++++++----------------
>  drivers/cxl/cxl.h         |   5 ++
>  2 files changed, 92 insertions(+), 60 deletions(-)
> 
> diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
> index 29d3809ab2bb..71c01447b234 100644
> --- a/drivers/cxl/core/region.c
> +++ b/drivers/cxl/core/region.c
> @@ -2928,28 +2928,66 @@ static bool has_spa_to_hpa(struct cxl_root_decoder *cxlrd)
>  	return cxlrd->ops && cxlrd->ops->spa_to_hpa;
>  }
>  
> -u64 cxl_dpa_to_hpa(struct cxl_region *cxlr, const struct cxl_memdev *cxlmd,
> -		   u64 dpa)
> +__mock_export u64 cxl_calculate_dpa_offset(u64 hpa_offset, u8 eiw, u16 eig)
>  {
> -	struct cxl_root_decoder *cxlrd = to_cxl_root_decoder(cxlr->dev.parent);
> -	u64 dpa_offset, hpa_offset, bits_upper, mask_upper, hpa;
> -	struct cxl_region_params *p = &cxlr->params;
> -	struct cxl_endpoint_decoder *cxled = NULL;
> -	u16 eig = 0;
> -	u8 eiw = 0;
> +	u64 dpa_offset, bits_lower, bits_upper, temp;
> +
> +	/*
> +	 * DPA offset: CXL Spec 3.2 Section 8.2.4.20.13
> +	 * Lower bits [IG+7:0] pass through unchanged
> +	 * (eiw < 8)
> +	 *	Per spec: DPAOffset[51:IG+8] = (HPAOffset[51:IG+IW+8] >> IW)
> +	 *	Clear the position bits to isolate upper section, then
> +	 *	reverse the left shift by eiw that occurred during DPA->HPA
> +	 * (eiw >= 8)
> +	 *	Per spec: DPAOffset[51:IG+8] = HPAOffset[51:IG+IW] / 3
> +	 *	Extract upper bits from the correct bit range and divide by 3
> +	 *	to recover the original DPA upper bits
> +	 */
> +	bits_lower = hpa_offset & GENMASK_ULL(eig + 7, 0);
> +	if (eiw < 8) {
> +		temp = hpa_offset &= ~((u64)GENMASK(eig + eiw + 8 - 1, 0));

Should this use GENMASK_ULL() instead of casting to u64?

> +		dpa_offset = temp >> eiw;
> +	} else {
> +		bits_upper = div64_u64(hpa_offset >> (eig + eiw), 3);
> +		dpa_offset = bits_upper << (eig + 8);
> +	}
> +	dpa_offset |= bits_lower;
> +
> +	return dpa_offset;
> +}
> +
> +__mock_export int cxl_calculate_position(u64 hpa_offset, u8 eiw, u16 eig)
> +{
> +	unsigned int ways = 0;
> +	u64 shifted, rem;
>  	int pos;
>  
> -	for (int i = 0; i < p->nr_targets; i++) {
> -		cxled = p->targets[i];
> -		if (cxlmd == cxled_to_memdev(cxled))
> -			break;
> +	/*
> +	 * Interleave position: CXL Spec 3.2 Section 8.2.4.20.13
> +	 * eiw < 8
> +	 *	Position is in the IW bits at HPA_OFFSET[IG+8+IW-1:IG+8].
> +	 *	Per spec "remove IW bits starting with bit position IG+8"
> +	 * eiw >= 8
> +	 *	Position is not explicitly stored in HPA_OFFSET bits. It is
> +	 *	derived from the modulo operation of the upper bits using
> +	 *	the total number of interleave ways.
> +	 */
> +	if (eiw < 8) {
> +		pos = (hpa_offset >> (eig + 8)) & GENMASK(eiw - 1, 0);
> +	} else {
> +		shifted = hpa_offset >> (eig + 8);
> +		eiw_to_ways(eiw, &ways);
> +		div64_u64_rem(shifted, ways, &rem);
> +		pos = rem;
>  	}
> -	if (!cxled || cxlmd != cxled_to_memdev(cxled))
> -		return ULLONG_MAX;
>  
> -	pos = cxled->pos;
> -	ways_to_eiw(p->interleave_ways, &eiw);
> -	granularity_to_eig(p->interleave_granularity, &eig);
> +	return pos;
> +}
> +
> +__mock_export u64 cxl_calculate_hpa_offset(u64 dpa_offset, int pos, u8 eiw, u16 eig)
> +{
> +	u64 mask_upper, hpa_offset, bits_upper;
>  
>  	/*
>  	 * The device position in the region interleave set was removed
> @@ -2961,9 +2999,6 @@ u64 cxl_dpa_to_hpa(struct cxl_region *cxlr, const struct cxl_memdev *cxlmd,
>  	 * 8.2.4.19.13 Implementation Note: Device Decode Logic
>  	 */
>  
> -	/* Remove the dpa base */
> -	dpa_offset = dpa - cxl_dpa_resource_start(cxled);
> -
>  	mask_upper = GENMASK_ULL(51, eig + 8);
>  
>  	if (eiw < 8) {
> @@ -2978,6 +3013,35 @@ u64 cxl_dpa_to_hpa(struct cxl_region *cxlr, const struct cxl_memdev *cxlmd,
>  	/* The lower bits remain unchanged */
>  	hpa_offset |= dpa_offset & GENMASK_ULL(eig + 7, 0);
>  
> +	return hpa_offset;
> +}
> +
> +u64 cxl_dpa_to_hpa(struct cxl_region *cxlr, const struct cxl_memdev *cxlmd,
> +		   u64 dpa)
> +{
> +	struct cxl_root_decoder *cxlrd = to_cxl_root_decoder(cxlr->dev.parent);
> +	struct cxl_region_params *p = &cxlr->params;
> +	struct cxl_endpoint_decoder *cxled = NULL;
> +	u64 dpa_offset, hpa_offset, hpa;
> +	u16 eig = 0;
> +	u8 eiw = 0;
> +	int pos;
> +
> +	for (int i = 0; i < p->nr_targets; i++) {
> +		cxled = p->targets[i];
> +		if (cxlmd == cxled_to_memdev(cxled))
> +			break;
> +	}
> +	if (!cxled || cxlmd != cxled_to_memdev(cxled))
> +		return ULLONG_MAX;

Maybe:

	for (int i = 0; i < p->nr_targets; i++) {
		if (cxlmd == cxled_to_memdev(p->targets[i])) {
			cxled = p->targets[i];
			break;
		}
	}

	if (!cxled)
		return ULLONG_MAX;

DJ

> +
> +	pos = cxled->pos;
> +	ways_to_eiw(p->interleave_ways, &eiw);
> +	granularity_to_eig(p->interleave_granularity, &eig);
> +
> +	dpa_offset = dpa - cxl_dpa_resource_start(cxled);
> +	hpa_offset = cxl_calculate_hpa_offset(dpa_offset, pos, eiw, eig);
> +
>  	/* Apply the hpa_offset to the region base address */
>  	hpa = hpa_offset + p->res->start + p->cache_size;
>  
> @@ -3010,8 +3074,6 @@ static int region_offset_to_dpa_result(struct cxl_region *cxlr, u64 offset,
>  	struct cxl_root_decoder *cxlrd = to_cxl_root_decoder(cxlr->dev.parent);
>  	struct cxl_endpoint_decoder *cxled;
>  	u64 hpa, hpa_offset, dpa_offset;
> -	u64 bits_upper, bits_lower;
> -	u64 shifted, rem, temp;
>  	u16 eig = 0;
>  	u8 eiw = 0;
>  	int pos;
> @@ -3033,50 +3095,15 @@ static int region_offset_to_dpa_result(struct cxl_region *cxlr, u64 offset,
>  	} else {
>  		hpa_offset = offset;
>  	}
> -	/*
> -	 * Interleave position: CXL Spec 3.2 Section 8.2.4.20.13
> -	 * eiw < 8
> -	 *	Position is in the IW bits at HPA_OFFSET[IG+8+IW-1:IG+8].
> -	 *	Per spec "remove IW bits starting with bit position IG+8"
> -	 * eiw >= 8
> -	 *	Position is not explicitly stored in HPA_OFFSET bits. It is
> -	 *	derived from the modulo operation of the upper bits using
> -	 *	the total number of interleave ways.
> -	 */
> -	if (eiw < 8) {
> -		pos = (hpa_offset >> (eig + 8)) & GENMASK(eiw - 1, 0);
> -	} else {
> -		shifted = hpa_offset >> (eig + 8);
> -		div64_u64_rem(shifted, p->interleave_ways, &rem);
> -		pos = rem;
> -	}
> +
> +	pos = cxl_calculate_position(hpa_offset, eiw, eig);
>  	if (pos < 0 || pos >= p->nr_targets) {
>  		dev_dbg(&cxlr->dev, "Invalid position %d for %d targets\n",
>  			pos, p->nr_targets);
>  		return -ENXIO;
>  	}
>  
> -	/*
> -	 * DPA offset: CXL Spec 3.2 Section 8.2.4.20.13
> -	 * Lower bits [IG+7:0] pass through unchanged
> -	 * (eiw < 8)
> -	 *	Per spec: DPAOffset[51:IG+8] = (HPAOffset[51:IG+IW+8] >> IW)
> -	 *	Clear the position bits to isolate upper section, then
> -	 *	reverse the left shift by eiw that occurred during DPA->HPA
> -	 * (eiw >= 8)
> -	 *	Per spec: DPAOffset[51:IG+8] = HPAOffset[51:IG+IW] / 3
> -	 *	Extract upper bits from the correct bit range and divide by 3
> -	 *	to recover the original DPA upper bits
> -	 */
> -	bits_lower = hpa_offset & GENMASK_ULL(eig + 7, 0);
> -	if (eiw < 8) {
> -		temp = hpa_offset &= ~((u64)GENMASK(eig + eiw + 8 - 1, 0));
> -		dpa_offset = temp >> eiw;
> -	} else {
> -		bits_upper = div64_u64(hpa_offset >> (eig + eiw), 3);
> -		dpa_offset = bits_upper << (eig + 8);
> -	}
> -	dpa_offset |= bits_lower;
> +	dpa_offset = cxl_calculate_dpa_offset(hpa_offset, eiw, eig);
>  
>  	/* Look-up and return the result: a memdev and a DPA */
>  	for (int i = 0; i < p->nr_targets; i++) {
> diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
> index 4fe3df06f57a..57590d131f75 100644
> --- a/drivers/cxl/cxl.h
> +++ b/drivers/cxl/cxl.h
> @@ -922,5 +922,10 @@ bool cxl_endpoint_decoder_reset_detected(struct cxl_port *port);
>  #define __mock static
>  #endif
>  
> +/* Unit test build overrides this to export, otherwise static */
> +#ifndef __mock_export
> +#define __mock_export static
> +#endif
> +
>  u16 cxl_gpf_get_dvsec(struct device *dev);
>  #endif /* __CXL_H__ */


  reply	other threads:[~2025-09-04 22:05 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-29  7:21 [PATCH v2 0/3] CXL: Add a loadable module for address translation Alison Schofield
2025-08-29  7:21 ` [PATCH v2 1/3] cxl/region: Refactor address translation funcs for testing Alison Schofield
2025-09-04 22:05   ` Dave Jiang [this message]
2025-09-09 17:31     ` Alison Schofield
2025-09-09 16:10   ` Jonathan Cameron
2025-09-09 17:45     ` Alison Schofield
2025-09-10 12:32       ` Jonathan Cameron
2025-08-29  7:21 ` [PATCH v2 2/3] cxl/acpi: Make the XOR calculations available " Alison Schofield
2025-09-04 23:21   ` Dave Jiang
2025-09-09 17:33     ` Alison Schofield
2025-08-29  7:21 ` [PATCH v2 3/3] cxl/test: Add cxl_translate module for address translation testing Alison Schofield
2025-09-04 23:24   ` Dave Jiang

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=6deca3ef-2284-433f-ae91-ce1ff9f7a8b7@intel.com \
    --to=dave.jiang@intel.com \
    --cc=alison.schofield@intel.com \
    --cc=dan.j.williams@intel.com \
    --cc=dave@stgolabs.net \
    --cc=ira.weiny@intel.com \
    --cc=jonathan.cameron@huawei.com \
    --cc=linux-cxl@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