This reverts commit 0e1c378fce66db833e08770d888dda1c5ec7936a.
---
src/lib/CMakeLists.txt | 1 +
src/lib/close_asic.c | 2 +-
src/lib/discover.c | 3 -
src/lib/free_maps.c | 44 ++++++++++
src/lib/read_vram.c | 218 ++++++++++++++++++++++++++++++++++++++++++++-----
src/umr.h | 14 +++-
6 files changed, 254 insertions(+), 28 deletions(-)
create mode 100644 src/lib/free_maps.c
diff --git a/src/lib/CMakeLists.txt b/src/lib/CMakeLists.txt
index a21fdf8eea2d..78d827ac1bf1 100644
--- a/src/lib/CMakeLists.txt
+++ b/src/lib/CMakeLists.txt
@@ -14,6 +14,7 @@ add_library(umrcore STATIC
discover.c
dump_ib.c
find_reg.c
+ free_maps.c
mmio.c
query_drm.c
read_sensor.c
diff --git a/src/lib/close_asic.c b/src/lib/close_asic.c
index a140409e617b..d532a11fa671 100644
--- a/src/lib/close_asic.c
+++ b/src/lib/close_asic.c
@@ -29,6 +29,7 @@
void umr_free_asic(struct umr_asic *asic)
{
int x;
+ umr_free_maps(asic);
if (asic->pci.mem != NULL) {
// free PCI mapping
pci_device_unmap_range(asic->pci.pdevice, asic->pci.mem, asic->pci.pdevice->regions[asic->pci.region].size);
@@ -56,7 +57,6 @@ void umr_close_asic(struct umr_asic *asic)
cond_close(asic->fd.vram);
cond_close(asic->fd.gpr);
cond_close(asic->fd.drm);
- cond_close(asic->fd.iova);
umr_free_asic(asic);
}
}
diff --git a/src/lib/discover.c b/src/lib/discover.c
index ff7950e4e6ba..dcc212fc39e4 100644
--- a/src/lib/discover.c
+++ b/src/lib/discover.c
@@ -222,8 +222,6 @@ struct umr_asic *umr_discover_asic(struct umr_options *options)
asic->fd.vram = open(fname, O_RDWR);
snprintf(fname, sizeof(fname)-1, "/sys/kernel/debug/dri/%d/amdgpu_gpr", asic->instance);
asic->fd.gpr = open(fname, O_RDWR);
- snprintf(fname, sizeof(fname)-1, "/sys/kernel/debug/dri/%d/amdgpu_iova", asic->instance);
- asic->fd.iova = open(fname, O_RDWR);
asic->fd.drm = -1; // default to closed
// if appending to the fd list remember to update close_asic() and discover_by_did()...
} else {
@@ -237,7 +235,6 @@ struct umr_asic *umr_discover_asic(struct umr_options *options)
asic->fd.vram = -1;
asic->fd.gpr = -1;
asic->fd.drm = -1;
- asic->fd.iova = -1;
}
if (options->use_pci) {
diff --git a/src/lib/free_maps.c b/src/lib/free_maps.c
new file mode 100644
index 000000000000..e1d27cb177f7
--- /dev/null
+++ b/src/lib/free_maps.c
@@ -0,0 +1,44 @@
+/*
+ * Copyright 2017 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Tom St Denis <tom.stdenis at amd.com>
+ *
+ */
+#include "umr.h"
+
+static void recurse_free(struct umr_map *map)
+{
+ if (map->left)
+ recurse_free(map->left);
+ if (map->right)
+ recurse_free(map->right);
+ free(map);
+}
+
+void umr_free_maps(struct umr_asic *asic)
+{
+ if (!asic->maps)
+ return;
+
+ recurse_free(asic->maps->maps);
+ free(asic->maps);
+ asic->maps = NULL;
+}
diff --git a/src/lib/read_vram.c b/src/lib/read_vram.c
index 3a327fb8e681..6e8f1f931895 100644
--- a/src/lib/read_vram.c
+++ b/src/lib/read_vram.c
@@ -25,6 +25,163 @@
#include "umrapp.h"
#include <inttypes.h>
+// find a mapping or create node for it
+static struct umr_map *find_map(struct umr_dma_maps *maps, uint64_t dma_addr, int create)
+{
+ struct umr_map *n = maps->maps, **nn;
+ uint64_t key;
+
+ // addresses aren't terribly random
+ // so if we use an identity function on the search
+ // key we'll end up with a really unbalanced tree
+ // so mix up address a bit to randomize keys
+ key = dma_addr ^ (dma_addr >> 11);
+
+ if (!n) {
+ maps->maps = calloc(1, sizeof(maps->maps[0]));
+ maps->maps->key = key;
+ return maps->maps;
+ }
+
+ while (n->dma_addr != dma_addr) {
+ if (key > n->key)
+ nn = &n->left;
+ else
+ nn = &n->right;
+
+ if (*nn) {
+ n = *nn;
+ } else {
+ if (!create) return NULL;
+
+ // add the new node
+ *nn = calloc(1, sizeof(maps->maps[0]));
+ (*nn)->key = key;
+ return *nn;
+ }
+ }
+
+ return n;
+}
+
+// insert/replace mapping in array
+static int insert_map(struct umr_dma_maps *maps,
+ uint64_t dma_addr, uint64_t phys_addr, int valid)
+{
+ struct umr_map *map = find_map(maps, dma_addr, valid);
+
+ // don't add a new node if we're marking it invalid
+ if (map) {
+ if (valid) {
+ map->dma_addr = dma_addr;
+ map->phys_addr = phys_addr;
+ map->valid = valid;
+ } else {
+ struct umr_map *tmap = NULL;
+
+ // if marking invalid see if we can prune
+ // the tree a little if the node we're marking
+ // as invalid only has one child
+ if (map->left == NULL && map->right) {
+ tmap = map->right;
+ *map = *(map->right);
+ } else if (map->right == NULL && map->left) {
+ tmap = map->left;
+ *map = *(map->left);
+ }
+ if (tmap)
+ free(tmap);
+ }
+ }
+
+ return 0;
+}
+
+static int check_trace = 0;
+
+// try to convert a DMA address to physical via trace
+static uint64_t dma_to_phys(struct umr_asic *asic, uint64_t dma_addr)
+{
+ struct umr_map *map = find_map(asic->maps, dma_addr, 0);
+
+ if (map == NULL)
+ return dma_addr;
+
+ if (map->valid)
+ return map->phys_addr;
+ else
+ return map->dma_addr;
+}
+
+static int parse_trace(struct umr_asic *asic)
+{
+ FILE *f;
+ uint64_t d, p;
+ char *s, buf[512];
+ int err = -1, valid;
+ struct umr_dma_maps *maps = asic->maps;
+
+ if (!check_trace) {
+ check_trace = 1;
+ f = fopen("/sys/kernel/debug/tracing/events/ttm/ttm_dma_map/enable", "r");
+ if (!f) {
+ fprintf(stderr, "[WARNING]: kernel does not support TTM mapping trace, please update kernel\n");
+ } else {
+ fgets(buf, sizeof(buf)-1, f);
+ if (sscanf(buf, "%"SCNu64, &d) == 1) {
+ if (d != 1) {
+ fprintf(stderr,
+ "[WARNING]: ttm_dma_map trace is not enabled, VM decoding may fail!\n"
+ "[WARNING]: Enable with: 'echo 1 > /sys/kernel/debug/tracing/events/ttm/ttm_dma_map/enable'\n"
+ "[WARNING]: Enable with: 'echo 1 > /sys/kernel/debug/tracing/events/ttm/ttm_dma_unmap/enable'\n");
+ }
+ } else {
+ fprintf(stderr, "[ERROR]: could not read ttm_dma_map enable file\n");
+ }
+ fclose(f);
+ }
+ }
+
+ // try to open ~/trace first
+ snprintf(buf, sizeof(buf)-1, "%s/trace", getenv("HOME"));
+ f = fopen(buf, "r");
+ if (!f)
+ f = fopen("/sys/kernel/debug/tracing/trace", "r");
+ if (!f)
+ goto error;
+
+ while (fgets(buf, sizeof(buf)-1, f)) {
+ valid = -1;
+
+ s = strstr(buf, "ttm_dma_map");
+ if (s) {
+ s += strlen("ttm_dma_map");
+ valid = 1;
+ } else {
+ s = strstr(buf, "ttm_dma_unmap");
+ if (s) {
+ s += strlen("ttm_dma_unmap");
+ valid = 0;
+ }
+ }
+
+ if (valid != -1) {
+ s = strstr(s, asic->options.pci.name);
+ if (s) {
+ s += strlen(asic->options.pci.name) + 2;
+ if (sscanf(s, "0x%"SCNx64" => 0x%"SCNx64, &d, &p) == 2) {
+ if (insert_map(maps, d, p, valid))
+ goto error;
+ }
+ }
+ }
+ }
+ err = 0;
+error:
+ fclose(f);
+ return err;
+}
+
static void access_vram_via_mmio(struct umr_asic *asic, uint64_t address, uint32_t size, void *dst, int write_en)
{
uint32_t MM_INDEX, MM_INDEX_HI, MM_DATA;
@@ -60,27 +217,33 @@ static void access_vram_via_mmio(struct umr_asic *asic, uint64_t address, uint32
#define DEBUG(...)
#endif
-static int umr_access_sram(struct umr_asic *asic, uint64_t address, uint32_t size, void *dst, int write_en)
+static int umr_access_sram(uint64_t address, uint32_t size, void *dst, int write_en)
{
- DEBUG("Reading physical sys addr: 0x%llx\n", (unsigned long long)address);
+ int fd;
- if (asic->fd.iova < 0) {
- fprintf(stderr, "[ERROR]: amdgpu_iova not open, please update your kernel\n");
- return -1;
- }
+ DEBUG("Reading physical sys addr: 0x%llx\n", (unsigned long long)address);
- lseek(asic->fd.iova, address, SEEK_SET);
- if (write_en == 0) {
- memset(dst, 0xFF, size);
- if (read(asic->fd.iova, dst, size) != size) {
- return -1;
- }
- } else {
- if (write(asic->fd.iova, dst, size) != size) {
- return -1;
+ fd = open("/dev/fmem", O_RDWR);
+ if (fd < 0)
+ fd = open("/dev/mem", O_RDWR | O_DSYNC);
+ if (fd >= 0) {
+ lseek(fd, address, SEEK_SET);
+ if (write_en == 0) {
+ memset(dst, 0xFF, size);
+ if (read(fd, dst, size) != size) {
+ close(fd);
+ return -1;
+ }
+ } else {
+ if (write(fd, dst, size) != size) {
+ close(fd);
+ return -1;
+ }
}
+ close(fd);
+ return 0;
}
- return 0;
+ return -1;
}
@@ -226,7 +389,7 @@ static int umr_access_vram_vi(struct umr_asic *asic, uint32_t vmid,
return -1;
// compute starting address
- start_addr = pte_fields.page_base_addr + (address & 0xFFF);
+ start_addr = dma_to_phys(asic, pte_fields.page_base_addr) + (address & 0xFFF);
} else {
// depth == 0 == PTE only
pte_idx = (address >> 12);
@@ -251,7 +414,7 @@ static int umr_access_vram_vi(struct umr_asic *asic, uint32_t vmid,
return -1;
// compute starting address
- start_addr = pte_fields.page_base_addr + (address & 0xFFF);
+ start_addr = dma_to_phys(asic, pte_fields.page_base_addr) + (address & 0xFFF);
}
// read upto 4K from it
@@ -265,7 +428,7 @@ static int umr_access_vram_vi(struct umr_asic *asic, uint32_t vmid,
// allow destination to be NULL to simply use decoder
if (pdst) {
if (pte_fields.system) {
- if (umr_access_sram(asic, start_addr, chunk_size, pdst, write_en) < 0) {
+ if (umr_access_sram(start_addr, chunk_size, pdst, write_en) < 0) {
fprintf(stderr, "[ERROR]: Cannot access system ram, perhaps CONFIG_STRICT_DEVMEM is set in your kernel config?\n");
fprintf(stderr, "[ERROR]: Alternatively download and install /dev/fmem\n");
return -1;
@@ -544,7 +707,7 @@ static int umr_access_vram_ai(struct umr_asic *asic, uint32_t vmid,
return -1;
// compute starting address
- start_addr = pte_fields.page_base_addr + (address & 0xFFF);
+ start_addr = dma_to_phys(asic, pte_fields.page_base_addr) + (address & 0xFFF);
DEBUG("phys address to read from: %llx\n\n\n", (unsigned long long)start_addr);
} else {
// in AI+ the BASE_ADDR is treated like a PDE entry...
@@ -593,7 +756,7 @@ static int umr_access_vram_ai(struct umr_asic *asic, uint32_t vmid,
return -1;
// compute starting address
- start_addr = pte_fields.page_base_addr + (address & 0xFFF);
+ start_addr = dma_to_phys(asic, pte_fields.page_base_addr) + (address & 0xFFF);
}
// read upto 4K from it
@@ -609,7 +772,7 @@ static int umr_access_vram_ai(struct umr_asic *asic, uint32_t vmid,
// allow destination to be NULL to simply use decoder
if (pdst) {
if (pte_fields.system) {
- if (umr_access_sram(asic, start_addr, chunk_size, pdst, write_en) < 0) {
+ if (umr_access_sram(start_addr, chunk_size, pdst, write_en) < 0) {
fprintf(stderr, "[ERROR]: Cannot access system ram, perhaps CONFIG_STRICT_DEVMEM is set in your kernel config?\n");
fprintf(stderr, "[ERROR]: Alternatively download and install /dev/fmem\n");
return -1;
@@ -662,6 +825,17 @@ int umr_access_vram(struct umr_asic *asic, uint32_t vmid, uint64_t address, uint
return 0;
}
+ if (!asic->maps) {
+ asic->maps = calloc(1, sizeof(asic->maps[0]));
+ if (!asic->maps) {
+ fprintf(stderr, "[ERROR]: Out of memory building dma maps\n");
+ return -1;
+ }
+
+ if (parse_trace(asic))
+ return -1;
+ }
+
switch (asic->family) {
case FAMILY_SI:
case FAMILY_CIK:
diff --git a/src/umr.h b/src/umr.h
index a374a0d23301..3d2252e35608 100644
--- a/src/umr.h
+++ b/src/umr.h
@@ -210,6 +210,17 @@ struct umr_options {
} pci;
};
+struct umr_dma_maps {
+ struct umr_map {
+ uint64_t
+ dma_addr,
+ phys_addr,
+ key;
+ int valid;
+ struct umr_map *left, *right;
+ } *maps;
+};
+
struct umr_asic {
char *asicname;
int no_blocks;
@@ -232,8 +243,7 @@ struct umr_asic {
drm,
wave,
vram,
- gpr,
- iova;
+ gpr;
} fd;
struct {
struct pci_device *pdevice;
--
2.12.0
