* [PATCH 1/1] cxl/region: Fix the first aliased address miscalculation
@ 2025-03-17 7:01 Li Ming
2025-03-17 17:59 ` Alison Schofield
` (2 more replies)
0 siblings, 3 replies; 4+ messages in thread
From: Li Ming @ 2025-03-17 7:01 UTC (permalink / raw)
To: dave, jonathan.cameron, dave.jiang, alison.schofield,
vishal.l.verma, ira.weiny, dan.j.williams
Cc: linux-cxl, linux-kernel, Li Ming
In extended linear cache(ELC) case, cxl_port_get_spa_cache_alias() helps
to get the aliased address of a SPA, it considers the first address in
CXL memory range is "region start + region cache size + 1", but it
should be "region start + region cache size".
So if a SPA is equal to "region start + region cache size", its aliased
address should be "SPA - region cache size".
Signed-off-by: Li Ming <ming.li@zohomail.com>
---
drivers/cxl/core/region.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
index 6d8bdb53f258..c3f4dc244df7 100644
--- a/drivers/cxl/core/region.c
+++ b/drivers/cxl/core/region.c
@@ -3460,7 +3460,7 @@ u64 cxl_port_get_spa_cache_alias(struct cxl_port *endpoint, u64 spa)
if (!p->cache_size)
return ~0ULL;
- if (spa > p->res->start + p->cache_size)
+ if (spa >= p->res->start + p->cache_size)
return spa - p->cache_size;
return spa + p->cache_size;
--
2.34.1
^ permalink raw reply related [flat|nested] 4+ messages in thread* Re: [PATCH 1/1] cxl/region: Fix the first aliased address miscalculation
2025-03-17 7:01 [PATCH 1/1] cxl/region: Fix the first aliased address miscalculation Li Ming
@ 2025-03-17 17:59 ` Alison Schofield
2025-03-20 18:26 ` Ira Weiny
2025-03-20 22:09 ` Dave Jiang
2 siblings, 0 replies; 4+ messages in thread
From: Alison Schofield @ 2025-03-17 17:59 UTC (permalink / raw)
To: Li Ming
Cc: dave, jonathan.cameron, dave.jiang, vishal.l.verma, ira.weiny,
dan.j.williams, linux-cxl, linux-kernel
On Mon, Mar 17, 2025 at 03:01:24PM +0800, Li Ming wrote:
> In extended linear cache(ELC) case, cxl_port_get_spa_cache_alias() helps
> to get the aliased address of a SPA, it considers the first address in
> CXL memory range is "region start + region cache size + 1", but it
> should be "region start + region cache size".
>
> So if a SPA is equal to "region start + region cache size", its aliased
> address should be "SPA - region cache size".
Nice catch Ming!
Reviewed-by: Alison Schofield <alison.schofield@intel.com>
>
> Signed-off-by: Li Ming <ming.li@zohomail.com>
> ---
> drivers/cxl/core/region.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
> index 6d8bdb53f258..c3f4dc244df7 100644
> --- a/drivers/cxl/core/region.c
> +++ b/drivers/cxl/core/region.c
> @@ -3460,7 +3460,7 @@ u64 cxl_port_get_spa_cache_alias(struct cxl_port *endpoint, u64 spa)
> if (!p->cache_size)
> return ~0ULL;
>
> - if (spa > p->res->start + p->cache_size)
> + if (spa >= p->res->start + p->cache_size)
> return spa - p->cache_size;
>
> return spa + p->cache_size;
> --
> 2.34.1
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH 1/1] cxl/region: Fix the first aliased address miscalculation
2025-03-17 7:01 [PATCH 1/1] cxl/region: Fix the first aliased address miscalculation Li Ming
2025-03-17 17:59 ` Alison Schofield
@ 2025-03-20 18:26 ` Ira Weiny
2025-03-20 22:09 ` Dave Jiang
2 siblings, 0 replies; 4+ messages in thread
From: Ira Weiny @ 2025-03-20 18:26 UTC (permalink / raw)
To: Li Ming, dave, jonathan.cameron, dave.jiang, alison.schofield,
vishal.l.verma, ira.weiny, dan.j.williams
Cc: linux-cxl, linux-kernel, Li Ming
Li Ming wrote:
> In extended linear cache(ELC) case, cxl_port_get_spa_cache_alias() helps
> to get the aliased address of a SPA, it considers the first address in
> CXL memory range is "region start + region cache size + 1", but it
> should be "region start + region cache size".
>
> So if a SPA is equal to "region start + region cache size", its aliased
> address should be "SPA - region cache size".
>
> Signed-off-by: Li Ming <ming.li@zohomail.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
[snip]
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH 1/1] cxl/region: Fix the first aliased address miscalculation
2025-03-17 7:01 [PATCH 1/1] cxl/region: Fix the first aliased address miscalculation Li Ming
2025-03-17 17:59 ` Alison Schofield
2025-03-20 18:26 ` Ira Weiny
@ 2025-03-20 22:09 ` Dave Jiang
2 siblings, 0 replies; 4+ messages in thread
From: Dave Jiang @ 2025-03-20 22:09 UTC (permalink / raw)
To: Li Ming, dave, jonathan.cameron, alison.schofield, vishal.l.verma,
ira.weiny, dan.j.williams
Cc: linux-cxl, linux-kernel
On 3/17/25 12:01 AM, Li Ming wrote:
> In extended linear cache(ELC) case, cxl_port_get_spa_cache_alias() helps
> to get the aliased address of a SPA, it considers the first address in
> CXL memory range is "region start + region cache size + 1", but it
> should be "region start + region cache size".
>
> So if a SPA is equal to "region start + region cache size", its aliased
> address should be "SPA - region cache size".
>
> Signed-off-by: Li Ming <ming.li@zohomail.com>
Applied to cxl/next
> ---
> drivers/cxl/core/region.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
> index 6d8bdb53f258..c3f4dc244df7 100644
> --- a/drivers/cxl/core/region.c
> +++ b/drivers/cxl/core/region.c
> @@ -3460,7 +3460,7 @@ u64 cxl_port_get_spa_cache_alias(struct cxl_port *endpoint, u64 spa)
> if (!p->cache_size)
> return ~0ULL;
>
> - if (spa > p->res->start + p->cache_size)
> + if (spa >= p->res->start + p->cache_size)
> return spa - p->cache_size;
>
> return spa + p->cache_size;
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2025-03-20 22:09 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-03-17 7:01 [PATCH 1/1] cxl/region: Fix the first aliased address miscalculation Li Ming
2025-03-17 17:59 ` Alison Schofield
2025-03-20 18:26 ` Ira Weiny
2025-03-20 22:09 ` Dave Jiang
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox