ACPICA/IORT: Correct the comment for id_count
Subsystems (3)
Hanjun Guo
maintainerRFC PATCH 1/2 ACPICA/IORT: Correct the comment for id_count
In IORT spec
(http://infocenter.arm.com/help/topic/com.arm.doc.den0049d/DEN0049D_IO_Remapping_Table.pdf),id_num means Number of IDs minus one, update the comment.
---
This patch just for comments, needs to be upstream in ACPICA first.
include/acpi/actbl2.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/acpi/actbl2.h +1 -1
@@ -104,7 +104,7 @@ enum acpi_iort_node_type {struct acpi_iort_id_mapping {u32 input_base;/* Lowest value in input range */ -u32 id_count;/* Number of IDs */ +u32 id_count;/* Number of IDs in the range minus one */ u32 output_base;/* Lowest value in output range */ u32 output_reference;/* A reference to the output node */ u32 flags;
Signature · 1 line
1.7.12.4
Hanjun Guo
maintainerRFC PATCH 2/2 ACPI/IORT: Workaround for IORT ID count "minus one" issue
The IORT spec [0] says Number of IDs = The number of IDs in the range minus one, it is confusing but it was written down in the first version of the IORT spec. But the IORT ID mapping function iort_id_map() did something wrong from the start, which bails out if:
the request ID >= the input base + number of IDs
This is wrong because it ignored the "minus one", and breaks some valid usecases such as ID mapping to contain single device mapping without single mapping flag set.
Pankaj Bansal proposed a solution to fix the issue [1], which bails out if:
the request ID > the input base + number of IDs
This works as the spec defined, unfortunately some firmware didn't minus one for the number of IDs in the range, and the propoased solution will break those systems in this way:
PCI hostbridge mapping entry 1:
Input base: 0x1000 ID Count: 0x100 Output base: 0x1000 Output reference: 0xC4 //ITS reference
PCI hostbridge mapping entry 2:
Input base: 0x1100 ID Count: 0x100 Output base: 0x2000 Output reference: 0xD4 //ITS reference
Two mapping entries which the second entry's Input base = the first entry's Input base + ID count, so for requester ID 0x1100 will map to ITS 0xC4 not 0xD4 if we update '>=' to '>'.
So introduce a workaround to match the IORT's OEM information for the broken firmware, also update the logic of the ID mapping for firmwares report the number of IDs as the IORT spec defined, to make the code compatible for both kinds of system.
I checked the ACPI tables in the tianocore/edk2-platforms [2], only HiSilicon HIP07/08 did wrong, so just add HIP07/08 to the workaround info table, if we break other platforms, we can add that later.
[0]: http://infocenter.arm.com/help/topic/com.arm.doc.den0049d/DEN0049D_IO_Remapping_Table.pdf
[1]: https://patchwork.kernel.org/patch/11292823/
[2]: https://github.com/tianocore/edk2-platforms
--- drivers/acpi/arm64/iort.c | 54 ++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 51 insertions(+), 3 deletions(-)
drivers/acpi/arm64/iort.c +51 -3
@@ -298,6 +298,41 @@ static acpi_status iort_match_node_callback(struct acpi_iort_node *node,return status; } +struct iort_workaround_oem_info { +char oem_id[ACPI_OEM_ID_SIZE +1 ]; +char oem_table_id[ACPI_OEM_TABLE_ID_SIZE +1 ]; +u32 oem_revision; +}; + +static bool apply_id_count_workaround; + +static struct iort_workaround_oem_info wa_info[]__initdata = { + { + .oem_id ="HISI " , + .oem_table_id ="HIP07 " , + .oem_revision =0 , + }, { + .oem_id ="HISI " , + .oem_table_id ="HIP08 " , + .oem_revision =0 , + } +}; + +static void __init +iort_check_id_count_workaround (struct acpi_table_header *tbl) +{ +int i; + +for (i =0 ; i <ARRAY_SIZE (wa_info); i++) { +if (!memcmp (wa_info[i].oem_id, tbl->oem_id,ACPI_OEM_ID_SIZE ) && + !memcmp (wa_info[i].oem_table_id, tbl->oem_table_id,ACPI_OEM_TABLE_ID_SIZE ) && + wa_info[i].oem_revision == tbl->oem_revision) { + apply_id_count_workaround = true; +break ; + } + } +} +static int iort_id_map (struct acpi_iort_id_mapping *map,u8 type,u32 rid_in,u32 *rid_out) {
@@ -314,9 +349,21 @@ static int iort_id_map(struct acpi_iort_id_mapping *map, u8 type, u32 rid_in,return -ENXIO ; } -if (rid_in < map->input_base || - (rid_in >= map->input_base + map->id_count)) -return -ENXIO ; +/* +* IORT spec says Number of IDs = The number of IDs in the range minus +* one, but the IORT code ingored the "minus one", and some firmware +* did that too, so apply a workaround here to keep compatible with +* both new and old versions of the firmware. +*/ +if (apply_id_count_workaround) { +if (rid_in < map->input_base || + (rid_in >= map->input_base + map->id_count)) +return -ENXIO ; + }else { +if (rid_in < map->input_base || + (rid_in > map->input_base + map->id_count)) +return -ENXIO ; + } *rid_out = map->output_base + (rid_in - map->input_base);return 0 ;
@@ -1631,5 +1678,6 @@ void __init acpi_iort_init(void)return ; } +iort_check_id_count_workaround (iort_table);iort_init_platform_devices (); }
Signature · 1 line
1.7.12.4
Re: [RFC PATCH 2/2] ACPI/IORT: Workaround for IORT ID count "minus one" issue
Can we put a warning of the sort to update the ACPI table with correct values?
eventually i would like that this workaround be removed from linux, when sufficient time has passed and all the platforms' ACPI tables have been updated.?
From: Hanjun Guo <[email protected]>
Sent: Monday, December 23, 2019 2:53 PM
To: Lorenzo Pieralisi <[email protected]>; Sudeep Holla <[email protected]>; Rafael J. Wysocki <[email protected]>; Pankaj Bansal <[email protected]>; Erik Schmauss <[email protected]>
Subject: [RFC PATCH 2/2] ACPI/IORT: Workaround for IORT ID count "minus one" issue
The IORT spec [0] says Number of IDs = The number of IDs in the range minus
one, it is confusing but it was written down in the first version of the
IORT spec. But the IORT ID mapping function iort_id_map() did something
wrong from the start, which bails out if:
the request ID >= the input base + number of IDs
This is wrong because it ignored the "minus one", and breaks some valid
usecases such as ID mapping to contain single device mapping without
single mapping flag set.
Pankaj Bansal proposed a solution to fix the issue [1], which bails
out if:
the request ID > the input base + number of IDs
This works as the spec defined, unfortunately some firmware didn't
minus one for the number of IDs in the range, and the propoased
solution will break those systems in this way:
PCI hostbridge mapping entry 1:
Input base: 0x1000
ID Count: 0x100
Output base: 0x1000
Output reference: 0xC4 //ITS reference
PCI hostbridge mapping entry 2:
Input base: 0x1100
ID Count: 0x100
Output base: 0x2000
Output reference: 0xD4 //ITS reference
Two mapping entries which the second entry's Input base = the first
entry's Input base + ID count, so for requester ID 0x1100 will map
to ITS 0xC4 not 0xD4 if we update '>=' to '>'.
So introduce a workaround to match the IORT's OEM information for
the broken firmware, also update the logic of the ID mapping for
firmwares report the number of IDs as the IORT spec defined, to
make the code compatible for both kinds of system.
I checked the ACPI tables in the tianocore/edk2-platforms [2], only
HiSilicon HIP07/08 did wrong, so just add HIP07/08 to the workaround
info table, if we break other platforms, we can add that later.
---
drivers/acpi/arm64/iort.c | 54 ++++++++++++++++++++++++++++++++++++++++++++---
1 file changed, 51 insertions(+), 3 deletions(-)
drivers/acpi/arm64/iort.c +0 -0
index 33f7198..112b1b0 100644
--- a/drivers/acpi/arm64/iort.c
+++ b/drivers/acpi/arm64/iort.c
@@ -298,6 +298,41 @@ static acpi_status iort_match_node_callback(struct acpi_iort_node *node,
return status;
}
+struct iort_workaround_oem_info {
+ char oem_id[ACPI_OEM_ID_SIZE + 1];
+ char oem_table_id[ACPI_OEM_TABLE_ID_SIZE + 1];
+ u32 oem_revision;
+};
+
+static bool apply_id_count_workaround;
+
+static struct iort_workaround_oem_info wa_info[] __initdata = {
+ {
+ .oem_id = "HISI ",
+ .oem_table_id = "HIP07 ",
+ .oem_revision = 0,
+ }, {
+ .oem_id = "HISI ",
+ .oem_table_id = "HIP08 ",
+ .oem_revision = 0,
+ }
+};
+
+static void __init
+iort_check_id_count_workaround(struct acpi_table_header *tbl)
+{
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(wa_info); i++) {
+ if (!memcmp(wa_info[i].oem_id, tbl->oem_id, ACPI_OEM_ID_SIZE) &&
+ !memcmp(wa_info[i].oem_table_id, tbl->oem_table_id, ACPI_OEM_TABLE_ID_SIZE) &&
+ wa_info[i].oem_revision == tbl->oem_revision) {
+ apply_id_count_workaround = true;
+ break;
+ }
+ }
+}
+
static int iort_id_map(struct acpi_iort_id_mapping *map, u8 type, u32 rid_in,
u32 *rid_out)
{
@@ -314,9 +349,21 @@ static int iort_id_map(struct acpi_iort_id_mapping *map, u8 type, u32 rid_in,
return -ENXIO;
}
- if (rid_in < map->input_base ||
- (rid_in >= map->input_base + map->id_count))
- return -ENXIO;
+ /*
+ * IORT spec says Number of IDs = The number of IDs in the range minus
+ * one, but the IORT code ingored the "minus one", and some firmware
+ * did that too, so apply a workaround here to keep compatible with
+ * both new and old versions of the firmware.
+ */
+ if (apply_id_count_workaround) {
+ if (rid_in < map->input_base ||
+ (rid_in >= map->input_base + map->id_count))
+ return -ENXIO;
+ } else {
+ if (rid_in < map->input_base ||
+ (rid_in > map->input_base + map->id_count))
+ return -ENXIO;
+ }
*rid_out = map->output_base + (rid_in - map->input_base);
return 0;
@@ -1631,5 +1678,6 @@ void __init acpi_iort_init(void)
return;
}
+ iort_check_id_count_workaround(iort_table);
iort_init_platform_devices();
}
Signature · 1 line
1.7.12.4
Hanjun Guo
maintainerRe: [RFC PATCH 2/2] ACPI/IORT: Workaround for IORT ID count "minus one" issue
Quoted 2 lines · Pankaj Bansal
Can we put a warning of the sort to update the> ACPI table with correct values?
Good point, I will add the firmware bug warning
in iort_check_id_count_workaround().
Quoted 3 lines
eventually i would like that this workaround be removed from linux, when sufficient time has passed and all the platforms' ACPI tables have been updated.?
I agree with you but it will take much longer time than expected.
Thanks
Hanjun
RE: [RFC PATCH 2/2] ACPI/IORT: Workaround for IORT ID count "minus one" issue
In reply to Hanjun GuoHi Hanjun,
Quoted 110 lines
-----Original Message-----
From: Linuxarm [mailto:[email protected]] On Behalf Of Hanjun Guo
Sent: 23 December 2019 09:23
To: Lorenzo Pieralisi <[email protected]>; Sudeep Holla <[email protected]>; Rafael J. Wysocki <[email protected]>; Pankaj Bansal <[email protected]>; Erik Schmauss <[email protected]>
Cc: [email protected]; Linuxarm <[email protected]>; [email protected]
Subject: [RFC PATCH 2/2] ACPI/IORT: Workaround for IORT ID count "minus one" issueThe IORT spec [0] says Number of IDs = The number of IDs in the range minus one, it is confusing but it was written down in the first version of the IORT spec. But the IORT ID mapping function
iort_id_map()did something wrong from the start, which bails out if:the request ID >= the input base + number of IDs
This is wrong because it ignored the "minus one", and breaks some valid usecases such as ID mapping to contain single device mapping without single mapping flag set.
Pankaj Bansal proposed a solution to fix the issue [1], which bails out if:
the request ID > the input base + number of IDs
This works as the spec defined, unfortunately some firmware didn't minus one for the number of IDs in the range, and the propoased solution will break those systems in this way:
PCI hostbridge mapping entry 1:
Input base: 0x1000 ID Count: 0x100 Output base: 0x1000 Output reference: 0xC4 //ITS referencePCI hostbridge mapping entry 2:
Input base: 0x1100 ID Count: 0x100 Output base: 0x2000 Output reference: 0xD4 //ITS referenceTwo mapping entries which the second entry's Input base = the first entry's Input base + ID count, so for requester ID
0x1100will map to ITS0xC4not0xD4if we update '>=' to '>'.So introduce a workaround to match the IORT's OEM information for the broken firmware, also update the logic of the ID mapping for firmwares report the number of IDs as the IORT spec defined, to make the code compatible for both kinds of system.
I checked the ACPI tables in the tianocore/edk2-platforms [2], only HiSilicon HIP07/08 did wrong, so just add HIP07/08 to the workaround info table, if we break other platforms, we can add that later.
[0]:
http://infocenter.arm.com/help/topic/com.arm.doc.den0049d/DEN0049D_IO_
Remapping_Table.pdf
[1]: https://patchwork.kernel.org/patch/11292823/
[2]: https://github.com/tianocore/edk2-platformsCc Pankaj Bansal <[email protected]>Cc Lorenzo Pieralisi <[email protected]>Signed-off-by Hanjun Guo <[email protected]>--- drivers/acpi/arm64/iort.c | 54 ++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 51 insertions(+), 3 deletions(-)drivers/acpi/arm64/iort.c +27
-0@@ -298,6 +298,41 @@ static acpi_status iort_match_node_callback(structacpi_iort_node *node, return status; } +struct iort_workaround_oem_info { +char oem_id[ACPI_OEM_ID_SIZE +1 ]; +char oem_table_id[ACPI_OEM_TABLE_ID_SIZE +1 ]; +u32 oem_revision; +}; + +static bool apply_id_count_workaround; + +static struct iort_workaround_oem_info wa_info[]__initdata = { + { + .oem_id ="HISI " , + .oem_table_id ="HIP07 " , + .oem_revision =0 , + }, { + .oem_id ="HISI " , + .oem_table_id ="HIP08 " , + .oem_revision =0 , + } +}; + +static void __init +iort_check_id_count_workaround (struct acpi_table_header *tbl) +{ +int i; + +for (i =0 ; i <ARRAY_SIZE (wa_info); i++) { + if (!memcmp(wa_info[i].oem_id, tbl->oem_id, ACPI_OEM_ID_SIZE)&&
drivers/acpi/arm64/iort.c +1
-0+ !memcmp(wa_info[i].oem_table_id, tbl->oem_table_id,
ACPI_OEM_TABLE_ID_SIZE) &&
drivers/acpi/arm64/iort.c static acpi_status iort_match_node_callback(struct acpi_iort_node *node, · line 335
Quoted 7 lines
drivers/acpi/arm64/iort.c +7
-0+ wa_info[i].oem_revision == tbl->oem_revision) { + apply_id_count_workaround = true; +break ; + } + } +} +
Can we get rid of the above and instead use acpi_match_platform_list() ? Please take a look at the pmcg_plat_info used for the HIP08 SMMUv3 PMCG erratum.
Thanks,
Shameer
Quoted 43 lines
static int iort_id_map(struct
acpi_iort_id_mapping*map, u8 type, u32rid_in,drivers/acpi/arm64/iort.c +15
-3u32 *rid_out) {@@ -314,9 +349,21 @@ static int iort_id_map(struct acpi_iort_id_mapping *map, u8 type, u32 rid_in, return -ENXIO ; } -if (rid_in < map->input_base || - (rid_in >= map->input_base + map->id_count)) -return -ENXIO ; +/* +* IORT spec says Number of IDs = The number of IDs in the range minus +* one, but the IORT code ingored the "minus one", and some firmware +* did that too, so apply a workaround here to keep compatible with +* both new and old versions of the firmware. +*/ +if (apply_id_count_workaround) { +if (rid_in < map->input_base || + (rid_in >= map->input_base + map->id_count)) +return -ENXIO ; + }else { +if (rid_in < map->input_base || + (rid_in > map->input_base + map->id_count)) +return -ENXIO ; + } *rid_out = map->output_base + (rid_in - map->input_base);drivers/acpi/arm64/iort.c +1
-1return 0 ;@@ -1631,5 +1678,6 @@ void __init acpi_iort_init(void)return ; } +iort_check_id_count_workaround (iort_table);iort_init_platform_devices (); } --1.7.12.4
_______________________________________________
Linuxarm mailing list
[email protected]
http://hulk.huawei.com/mailman/listinfo/linuxarm
Hanjun Guo
maintainerRe: [RFC PATCH 2/2] ACPI/IORT: Workaround for IORT ID count "minus one" issue
Quoted 114 lines · Shameerali Kolothum Thodi
Hi Hanjun,
Quoted 112 lines
-----Original Message-----
From: Linuxarm [mailto:[email protected]] On Behalf Of Hanjun Guo
Sent: 23 December 2019 09:23
To: Lorenzo Pieralisi <[email protected]>; Sudeep Holla <[email protected]>; Rafael J. Wysocki <[email protected]>; Pankaj Bansal <[email protected]>; Erik Schmauss <[email protected]>
Cc: [email protected]; Linuxarm <[email protected]>; [email protected]
Subject: [RFC PATCH 2/2] ACPI/IORT: Workaround for IORT ID count "minus one" issueThe IORT spec [0] says Number of IDs = The number of IDs in the range minus one, it is confusing but it was written down in the first version of the IORT spec. But the IORT ID mapping function
iort_id_map()did something wrong from the start, which bails out if:the request ID >= the input base + number of IDs
This is wrong because it ignored the "minus one", and breaks some valid usecases such as ID mapping to contain single device mapping without single mapping flag set.
Pankaj Bansal proposed a solution to fix the issue [1], which bails out if:
the request ID > the input base + number of IDs
This works as the spec defined, unfortunately some firmware didn't minus one for the number of IDs in the range, and the propoased solution will break those systems in this way:
PCI hostbridge mapping entry 1:
Input base: 0x1000 ID Count: 0x100 Output base: 0x1000 Output reference: 0xC4 //ITS referencePCI hostbridge mapping entry 2:
Input base: 0x1100 ID Count: 0x100 Output base: 0x2000 Output reference: 0xD4 //ITS referenceTwo mapping entries which the second entry's Input base = the first entry's Input base + ID count, so for requester ID
0x1100will map to ITS0xC4not0xD4if we update '>=' to '>'.So introduce a workaround to match the IORT's OEM information for the broken firmware, also update the logic of the ID mapping for firmwares report the number of IDs as the IORT spec defined, to make the code compatible for both kinds of system.
I checked the ACPI tables in the tianocore/edk2-platforms [2], only HiSilicon HIP07/08 did wrong, so just add HIP07/08 to the workaround info table, if we break other platforms, we can add that later.
[0]:
http://infocenter.arm.com/help/topic/com.arm.doc.den0049d/DEN0049D_IO_
Remapping_Table.pdf
[1]: https://patchwork.kernel.org/patch/11292823/
[2]: https://github.com/tianocore/edk2-platformsCc Pankaj Bansal <[email protected]>Cc Lorenzo Pieralisi <[email protected]>Signed-off-by Hanjun Guo <[email protected]>--- drivers/acpi/arm64/iort.c | 54 ++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 51 insertions(+), 3 deletions(-)drivers/acpi/arm64/iort.c +27
-0@@ -298,6 +298,41 @@ static acpi_status iort_match_node_callback(structacpi_iort_node *node, return status; } +struct iort_workaround_oem_info { +char oem_id[ACPI_OEM_ID_SIZE +1 ]; +char oem_table_id[ACPI_OEM_TABLE_ID_SIZE +1 ]; +u32 oem_revision; +}; + +static bool apply_id_count_workaround; + +static struct iort_workaround_oem_info wa_info[]__initdata = { + { + .oem_id ="HISI " , + .oem_table_id ="HIP07 " , + .oem_revision =0 , + }, { + .oem_id ="HISI " , + .oem_table_id ="HIP08 " , + .oem_revision =0 , + } +}; + +static void __init +iort_check_id_count_workaround (struct acpi_table_header *tbl) +{ +int i; + +for (i =0 ; i <ARRAY_SIZE (wa_info); i++) { + if (!memcmp(wa_info[i].oem_id, tbl->oem_id, ACPI_OEM_ID_SIZE)&&
drivers/acpi/arm64/iort.c +1
-0+ !memcmp(wa_info[i].oem_table_id, tbl->oem_table_id,
ACPI_OEM_TABLE_ID_SIZE) &&drivers/acpi/arm64/iort.c +2
-0+ wa_info[i].oem_revision == tbl->oem_revision) { + apply_id_count_workaround = true;
Quoted 8 lines
drivers/acpi/arm64/iort.c static acpi_status iort_match_node_callback(struct acpi_iort_node *node, · line 335
Quoted 5 lines
drivers/acpi/arm64/iort.c +5
-0+break ; + } + } +} +Can we get rid of the above and instead use
acpi_match_platform_list()? Please take a look at thepmcg_plat_infoused for the HIP08 SMMUv3 PMCG erratum.
Thanks for the reminding, I noticed acpi_match_platform_list() before but I thought it was a little overkill (get the table header for each OEM info), I will take a look further to see if I can consolidate the OEM information retrieve.
Thanks
Hanjun
Quoted 17 lines · Hanjun Guo
In IORT spec
(http://infocenter.arm.com/help/topic/com.arm.doc.den0049d/DEN0049D_IO_Remapping_Table.pdf),id_nummeans Number of IDs minus one, update the comment.Signed-off-by Hanjun Guo <[email protected]>---
This patch just for comments, needs to be upstream in ACPICA first.
include/acpi/actbl2.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)include/acpi/actbl2.h +0
-0
include/acpi/actbl2.h enum acpi_iort_node_type { · line 107
Quoted 5 lines
include/acpi/actbl2.h +1
-1@@ -104,7 +104,7 @@ enum acpi_iort_node_type {struct acpi_iort_id_mapping {u32 input_base;/* Lowest value in input range */ -u32 id_count;/* Number of IDs */ +u32 id_count;/* Number of IDs in the range minus one */
The IORT spec also uses the term "Length" in the examples...
Quoted 4 lines
include/acpi/actbl2.h +0
-0u32 output_base;/* Lowest value in output range */ u32 output_reference;/* A reference to the output node */ u32 flags;
Hanjun Guo
maintainerQuoted 13 lines · John Garry
Quoted 12 lines · Hanjun Guo
In IORT spec
(http://infocenter.arm.com/help/topic/com.arm.doc.den0049d/DEN0049D_IO_Remapping_Table.pdf),id_nummeans Number of IDs minus one, update the comment.Signed-off-by Hanjun Guo <[email protected]>---
This patch just for comments, needs to be upstream in ACPICA first.
include/acpi/actbl2.h| 2 +-1 file changed, 1 insertion(+), 1 deletion(-)
Quoted 11 lines
Quoted 9 lines
include/acpi/actbl2.h +0
-0@@ -104,7 +104,7 @@ enum acpi_iort_node_type { struct acpi_iort_id_mapping {u32 input_base; /* Lowest value in input range */ - u32 id_count; /* Number of IDs */
+ u32 id_count; /* Number of IDs in the range minus one */The IORT spec also uses the term "Length" in the examples...
More confusing...