Fix warning

This is a major change in the SoC detection methodology.

Previous to this commit, the SoC (model, vendor and manufacturing
process) would only appear if the exact model was found in the LUT.
In other words, every supported SoC must be added manually to the LUT.
This would allow to show the precise SoC model and the manufacturing
process, which can only be hardcoded (i.e., it cannot be dynamically
determined).

This commit introduces guess_raw_soc_from_devtree, a new way of
inferring the SoC. This simply reads from the compatible file in the
device tree and tries to find a matching vendor to that string. If
there is a match, then we simply copy the model also from the
compatible string and show it directly.

This new implementation will show a less "precise" SoC name and
no manufacturing process because there would be no hardcoded value for
that SoC. However, it improves the scalability of the SoC detection
significantly because there is no longer a need to hardcode every
single SoC.

Lastly, there are some SoC vendors intentionally left outside the
scope of this function: For now I prefer to keep updating manually
the LUT for those to ensure the highest quality detection (e.g.,
showing precise SoC name and manufacturing process).

src/arm/soc.c

@@ -1067,6 +1067,64 @@ struct system_on_chip* guess_soc_from_devtree(struct system_on_chip* soc) {
   DT_END(dt, len)
 }
 
+// This function is different from the rest guess_soc_from_xxx, which try infering
+// the exact SoC model by matching some string against a list of known values.
+// On the other hand, this function will just try to infer the SoC vendor first by
+// matching the device tree vendor name (i.e., the first value, before the comma).
+// If that is successfull, then it also fills in the SoC name using the string from
+// the device tree.
+// The critical difference is that this function does not need a LUT to fill in the
+// SoC, it just needs to find a known vendor. On the other hand, the detection is
+// less powerful since we cannot get the manufacturing process, and the SoC name will
+// come directly from the device tree, meaning that it will likely be less precise.
+struct system_on_chip* guess_raw_soc_from_devtree(struct system_on_chip* soc) {
+  int num_vendors;
+  struct devtree** dt_vendors = get_devtree_compatible_struct(&num_vendors);
+  if (dt_vendors == NULL) {
+    return soc;
+  }
+
+  typedef struct {
+    char* compatible;
+    VENDOR soc_vendor;
+  } devtreeToVendor;
+
+  // https://elixir.bootlin.com/linux/v6.10.6/source/arch/arm64/boot/dts
+  // grep -oR --color -E 'compatible = ".*"' <soc_vendor> | cut -d '=' -f2 | cut -d ',' -f1 | tr -d '"' | sort | uniq -c | sort
+  // - The following vendors are not included because they dont seem to be present in dts:
+  // SOC_VENDOR_(KIRIN, KUNPENG, GOOGLE, AMPERE).
+  // - The commented vendors are not included intentionally, because I prefer updating its LUT manually.
+  devtreeToVendor socFromDevtree[] = {
+    // {"qcom",       SOC_VENDOR_SNAPDRAGON},
+    // {"samsung",    SOC_VENDOR_EXYNOS},
+    // {"brcm",       SOC_VENDOR_BROADCOM},
+    // {"apple",      SOC_VENDOR_APPLE},
+    // {"rockchip",   SOC_VENDOR_ROCKCHIP},
+    // {"nvidia",     SOC_VENDOR_NVIDIA},
+    {"mediatek",   SOC_VENDOR_MEDIATEK},
+    {"fsl",        SOC_VENDOR_NXP     },
+    {"nxp",        SOC_VENDOR_NXP     },
+    {"amlogic",    SOC_VENDOR_AMLOGIC },
+    {"marvell",    SOC_VENDOR_MARVELL },
+    {NULL,         SOC_VENDOR_UNKNOWN }
+  };
+
+  int index = 0;
+  while (socFromDevtree[index].compatible != 0x0) {
+    for (int i=0; i < num_vendors; i++) {
+      if (strcmp(socFromDevtree[index].compatible, dt_vendors[i]->vendor) == 0) {
+        fill_soc_raw(soc, dt_vendors[i]->model, socFromDevtree[index].soc_vendor);
+        printWarn("Your SoC is unsupported by cpufetch but could still be detected successfully. If you want to help improve the project, please paste the output of 'cpufetch --verbose' on https://github.com/Dr-Noob/cpufetch/issues");
+        return soc;
+      }
+    }
+    index++;
+  }
+
+  printWarn("guess_raw_soc_from_devtree: No device matched the list");
+  return soc;
+}
+
 struct system_on_chip* guess_soc_from_pci(struct system_on_chip* soc, struct cpuInfo* cpu) {
   struct pci_devices * pci = get_pci_devices();
   if (pci == NULL) {
@@ -1293,6 +1351,11 @@ struct system_on_chip* get_soc(struct cpuInfo* cpu) {
     if(soc->vendor == SOC_VENDOR_UNKNOWN) {
       soc = guess_soc_from_pci(soc, cpu);
     }
+    if (soc->vendor == SOC_VENDOR_UNKNOWN) {
+      // If we fall here it means all previous functions failed to detect the SoC.
+      // In such case, try with our last resort. If it also fails, we will just give up
+      soc = guess_raw_soc_from_devtree(soc);
+    }
   }
 #elif defined __APPLE__ || __MACH__
   soc = guess_soc_apple(soc);
@@ -1318,16 +1381,13 @@ struct system_on_chip* get_soc(struct cpuInfo* cpu) {
   soc->vendor = try_match_soc_vendor_name(processor_name_string);
   soc->model = SOC_MODEL_UNKNOWN;
   soc->process = UNKNOWN;
-
 #else
-
-  if(soc->model == SOC_MODEL_UNKNOWN) {
-    // raw_name might not be NULL, but if we were unable to find
-    // the exact SoC, just print "Unkwnown"
+  if(soc->raw_name == NULL) {
+    // We were unable to find the SoC, so just initialize raw_name
+    // with the unknown string
     soc->raw_name = emalloc(sizeof(char) * (strlen(STRING_UNKNOWN)+1));
     snprintf(soc->raw_name, strlen(STRING_UNKNOWN)+1, STRING_UNKNOWN);
   }
-
 #endif 
 
   return soc;

src/arm/socs.h

@@ -443,7 +443,7 @@ inline static VENDOR get_soc_vendor_from_soc(SOC soc) {
   else if(soc >= SOC_HISILICON_3620 && soc <= SOC_HISILICON_9000S) return SOC_VENDOR_KIRIN;
   else if(soc >= SOC_KUNPENG_920 && soc <= SOC_KUNPENG_930) return SOC_VENDOR_KUNPENG;
   else if(soc >= SOC_EXYNOS_3475 && soc <= SOC_EXYNOS_880) return SOC_VENDOR_EXYNOS;
-  else if(soc >= SOC_MTK_MT6893 && soc <= SOC_MTK_MT8796) return SOC_VENDOR_MEDIATEK;
+  else if(soc >= SOC_MTK_MT5327 && soc <= SOC_MTK_MT6833V) return SOC_VENDOR_MEDIATEK;
   else if(soc >= SOC_SNAPD_QSD8650 && soc <= SOC_SNAPD_SC8280XP) return SOC_VENDOR_SNAPDRAGON;
   else if(soc >= SOC_APPLE_M1 && soc <= SOC_APPLE_M3_MAX) return SOC_VENDOR_APPLE;
   else if(soc >= SOC_ALLWINNER_A10 && soc <= SOC_ALLWINNER_R328) return SOC_VENDOR_ALLWINNER;

src/common/ascii.h

@@ -445,6 +445,26 @@ $C1     .#########       |__________|######   |__________| \
 $C1   ###########         ___###########       __________  \
 $C1.##########           |__________|         |__________| "
 
+#define ASCII_SPACEMIT \
+"$C1                      :#: \
+$C1                   :####: \
+$C1                :#######: \
+$C1              :#########: \
+$C1              :#########: \
+$C1              :#######:   \
+$C1              :####:      \
+$C1              :#:         \
+$C1:##:             :#:      \
+$C1:####:         :###:      \
+$C1:#######:     :####:      \
+$C1:##########:  :###:       \
+$C1:###########: :#:         \
+$C1:###########:             \
+$C1 :##########:             \
+$C1    :#######:             \
+$C1       :####:             \
+$C1         :##:             "
+
 // --------------------- LONG LOGOS ------------------------- //
 #define ASCII_AMD_L \
 "$C1                                                              \
@@ -624,6 +644,7 @@ asciiL logo_ampere      = { ASCII_AMPERE,      50, 17, false, {C_FG_RED},
 asciiL logo_nxp         = { ASCII_NXP,         55,  8, false, {C_FG_YELLOW, C_FG_CYAN, C_FG_GREEN},           {C_FG_CYAN,    C_FG_WHITE}   };
 asciiL logo_amlogic     = { ASCII_AMLOGIC,     58,  8, false, {C_FG_BLUE},                                    {C_FG_BLUE,    C_FG_B_WHITE} };
 asciiL logo_marvell     = { ASCII_MARVELL,     56, 10, false, {C_FG_B_BLACK},                                 {C_FG_B_BLACK, C_FG_B_WHITE} };
+asciiL logo_spacemit    = { ASCII_SPACEMIT,    26, 18, false, {C_FG_B_GREEN},                                 {C_FG_B_GREEN, C_FG_B_WHITE} };
 
 // Long variants          | ----------------------------------------------------------------------------------------------------------------|
 asciiL logo_amd_l       = { ASCII_AMD_L,       62, 19, true,  {C_BG_WHITE, C_BG_GREEN},                       {C_FG_WHITE, C_FG_GREEN}     };

src/common/cpu.h

@@ -25,6 +25,7 @@ enum {
   CPU_VENDOR_RISCV,
   CPU_VENDOR_SIFIVE,
   CPU_VENDOR_THEAD,
+  CPU_VENDOR_SPACEMIT,
 // OTHERS
   CPU_VENDOR_UNKNOWN,
   CPU_VENDOR_INVALID

src/common/printer.c

@@ -20,6 +20,7 @@
   #include "../arm/uarch.h"
   #include "../arm/midr.h"
   #include "../arm/soc.h"
+  #include "../arm/socs.h"
   #include "../common/soc.h"
 #elif ARCH_RISCV
   #include "../riscv/riscv.h"
@@ -411,6 +412,8 @@ void choose_ascii_art(struct ascii* art, struct color** cs, struct terminal* ter
     art->art = &logo_allwinner;
   else if(art->vendor == SOC_VENDOR_SIPEED)
     art->art = &logo_sipeed;
+  else if(art->vendor == SOC_VENDOR_SPACEMIT)
+    art->art = &logo_spacemit;
   else
     art->art = &logo_riscv;
 #endif
@@ -892,6 +895,11 @@ bool print_cpufetch_arm(struct cpuInfo* cpu, STYLE s, struct color** cs, struct
   // https://github.com/Dr-Noob/cpufetch/pull/273
   // Hide manufacturing process
 #if !defined(_WIN32)
+  // In the case that the model is unknown but the vendor isn't (this is, when
+  // guess_raw_soc_from_devtree succeeded), do not show the manufacturing process
+  // (as it will be unknown)
+  if (cpu->soc->model != SOC_MODEL_UNKNOWN ||
+    (cpu->soc->model == SOC_MODEL_UNKNOWN && cpu->soc->vendor == SOC_VENDOR_UNKNOWN))
   setAttribute(art, ATTRIBUTE_TECHNOLOGY, manufacturing_process);
 #endif
 

src/common/soc.c

@@ -29,6 +29,7 @@ static char* soc_trademark_string[] = {
   [SOC_VENDOR_SIFIVE]     = "SiFive ",
   [SOC_VENDOR_STARFIVE]   = "StarFive ",
   [SOC_VENDOR_SIPEED]     = "Sipeed ",
+  [SOC_VENDOR_SPACEMIT]   = "SpacemiT ",
   // ARM & RISC-V
   [SOC_VENDOR_ALLWINNER]  = "Allwinner "
 };
@@ -79,6 +80,16 @@ void fill_soc(struct system_on_chip* soc, char* soc_name, SOC soc_model, int32_t
   }
 }
 
+void fill_soc_raw(struct system_on_chip* soc, char* soc_name, VENDOR vendor) {
+  soc->model = SOC_MODEL_UNKNOWN;
+  soc->vendor = vendor;
+  soc->process = UNKNOWN;
+
+  int len = strlen(soc_name) + strlen(soc_trademark_string[soc->vendor]) + 1;
+  soc->raw_name = emalloc(sizeof(char) * len);
+  sprintf(soc->raw_name, "%s%s", soc_trademark_string[soc->vendor], soc_name);
+}
+
 #ifdef _WIN32
 VENDOR try_match_soc_vendor_name(char* vendor_name)
 {

src/common/soc.h

@@ -33,6 +33,7 @@ enum {
   SOC_VENDOR_SIFIVE,
   SOC_VENDOR_STARFIVE,
   SOC_VENDOR_SIPEED,
+  SOC_VENDOR_SPACEMIT,
   // ARM & RISC-V
   SOC_VENDOR_ALLWINNER
 };
@@ -51,6 +52,7 @@ VENDOR get_soc_vendor(struct system_on_chip* soc);
 bool match_soc(struct system_on_chip* soc, char* raw_name, char* expected_name, char* soc_name, SOC soc_model, int32_t process);
 char* get_str_process(struct system_on_chip* soc);
 void fill_soc(struct system_on_chip* soc, char* soc_name, SOC soc_model, int32_t process);
+void fill_soc_raw(struct system_on_chip* soc, char* soc_name, VENDOR vendor);
 #ifdef _WIN32
 VENDOR try_match_soc_vendor_name(char* vendor_name);
 #endif

src/common/udev.c

@@ -361,3 +361,69 @@ char* get_devtree_compatible(int *filelen) {
 
   return buf;
 }
+
+// Returns a list of structs devtree, each containing both the vendor and the
+// model, coming from the compatible file from the device tree. In this
+// context, vendor refers to the first string of every entry and the model to
+// the second. For instance, given a compatible file with:
+// "str1,foo1.str2,foo2" (where . denotes the NULL byte, i.e., the separator),
+// then this function will return a list with two structs, the first one
+// containing str1 and foo1 and the other containing str2 and foo2.
+struct devtree** get_devtree_compatible_struct(int *num_vendors_ptr) {
+  int len;
+  char* dt = get_devtree_compatible(&len);
+  if (dt == NULL) {
+    return NULL;
+  }
+
+  int num_vendors = 0;
+  char* ptr = dt;
+
+  for (int ptrpos = 0; ptrpos < len; ptrpos = (ptr-dt)) {
+    ptr = memchr(ptr, '\0', len);
+    if (ptr == NULL) {
+      printBug("get_devtree_compatible_struct: Unable to find delimiter (1) (num_vendors=%d)", num_vendors);
+      return NULL;
+    }
+    ptr++;
+    num_vendors++;
+  }
+
+  struct devtree** vendors = emalloc(sizeof(struct devtree *) * num_vendors);
+  ptr = dt;
+
+  for (int ptrpos = 0, i = 0; ptrpos < len; ptrpos = (ptr-dt), i++) {
+    char* comma_ptr = strstr(ptr, ",");
+    if (comma_ptr == NULL) {
+      printBug("get_devtree_compatible_struct: Unable to find comma (num_vendors=%d)", num_vendors);
+      return NULL;
+    }
+    comma_ptr = comma_ptr-1; // Point right before comma
+
+    char* end_ptr = memchr(comma_ptr, '\0', len - ptrpos);
+    if (end_ptr == NULL) {
+      printBug("get_devtree_compatible_struct: Unable to find delimiter (2) (num_vendors=%d)", num_vendors);
+      return NULL;
+    }
+
+    int vendor_str_len = (comma_ptr-ptr)+1;
+    int model_str_len = (end_ptr-(comma_ptr+2))+1;
+
+    vendors[i] = emalloc(sizeof(struct devtree));
+    vendors[i]->vendor = ecalloc(vendor_str_len, sizeof(char));
+    vendors[i]->model = ecalloc(model_str_len, sizeof(char));
+
+    strncpy(vendors[i]->vendor, ptr, vendor_str_len);
+    strncpy(vendors[i]->model, comma_ptr+2, model_str_len);
+
+    ptr = memchr(ptr, '\0', len);
+    if (ptr == NULL) {
+      printBug("get_devtree_compatible_struct: Unable to find delimiter (3) (num_vendors=%d)", num_vendors);
+      return NULL;
+    }
+    ptr++; // Point right after delimiter
+  }
+
+  *num_vendors_ptr = num_vendors;
+  return vendors;
+}

src/common/udev.h

@@ -31,6 +31,11 @@
 #define _PATH_CACHE_MAX_LEN     200
 #define _PATH_PACKAGE_MAX_LEN   200
 
+struct devtree {
+  char* vendor;
+  char* model;
+};
+
 char* read_file(char* path, int* len);
 long get_max_freq_from_file(uint32_t core);
 long get_min_freq_from_file(uint32_t core);
@@ -44,5 +49,6 @@ int get_ncores_from_cpuinfo(void);
 char* get_field_from_cpuinfo(char* CPUINFO_FIELD);
 bool is_devtree_compatible(char* str);
 char* get_devtree_compatible(int *filelen);
+struct devtree** get_devtree_compatible_struct(int *num_vendors);
 
 #endif

src/riscv/riscv.c

@@ -101,8 +101,8 @@ struct extensions* get_extensions_from_str(char* str) {
     return ext;
   }
 
-  int len = strlen(str);
-  ext->str = ecalloc(len+1, sizeof(char));
+  int len = strlen(str)+1;
+  ext->str = emalloc(len * sizeof(char));
   strncpy(ext->str, str, sizeof(char) * len);
 
   // Code inspired in Linux kernel (riscv_fill_hwcap):
@@ -157,13 +157,12 @@ struct cpuInfo* get_cpu_info(void) {
   topo->cach = NULL;
   cpu->topo = topo;
 
-  char* cpuinfo_str = get_uarch_from_cpuinfo();
   char* ext_str = get_extensions_from_cpuinfo();
   cpu->hv = emalloc(sizeof(struct hypervisor));
   cpu->hv->present = false;
   cpu->ext = get_extensions_from_str(ext_str);
   if(cpu->ext->str != NULL && cpu->ext->mask == 0) return NULL;
-  cpu->arch = get_uarch_from_cpuinfo_str(cpuinfo_str, cpu);
+  cpu->arch = get_uarch(cpu);
   cpu->soc = get_soc(cpu);
   cpu->freq = get_frequency_info(0);
   cpu->peak_performance = get_peak_performance(cpu);

src/riscv/soc.c

@@ -38,6 +38,12 @@ bool match_sipeed(char* soc_name, struct system_on_chip* soc) {
   SOC_END
 }
 
+bool match_spacemit(char* soc_name, struct system_on_chip* soc) {
+  SOC_START
+  SOC_EQ(soc_name, "k1-x", "K1-X", SOC_SPACEMIT_K1X, soc, 22) // https://github.com/Dr-Noob/cpufetch/issues/286 https://www.spacemit.com/en/spacemit-x60-core/
+  SOC_END
+}
+
 struct system_on_chip* parse_soc_from_string(struct system_on_chip* soc) {
   char* raw_name = soc->raw_name;
 
@@ -50,6 +56,9 @@ struct system_on_chip* parse_soc_from_string(struct system_on_chip* soc) {
   if(match_sifive(raw_name, soc))
     return soc;
 
+  if(match_spacemit(raw_name, soc))
+    return soc;
+
   match_sipeed(raw_name, soc);
   return soc;
 }

src/riscv/socs.h

@@ -13,6 +13,8 @@ enum {
   SOC_ALLWINNER_D1H,
   // SIPEED
   SOC_SIPEED_LICHEEPI4A,
+  // SPACEMIT
+  SOC_SPACEMIT_K1X,
   // UNKNOWN
   SOC_MODEL_UNKNOWN
 };
@@ -22,6 +24,7 @@ inline static VENDOR get_soc_vendor_from_soc(SOC soc) {
   if(soc >= SOC_STARFIVE_VF2 && soc <= SOC_STARFIVE_VF2) return SOC_VENDOR_STARFIVE;
   if(soc >= SOC_ALLWINNER_D1H && soc <= SOC_ALLWINNER_D1H) return SOC_VENDOR_ALLWINNER;
   if(soc >= SOC_SIPEED_LICHEEPI4A && soc <= SOC_SIPEED_LICHEEPI4A) return SOC_VENDOR_SIPEED;
+  if(soc >= SOC_SPACEMIT_K1X && soc <= SOC_SPACEMIT_K1X) return SOC_VENDOR_SPACEMIT;
   return SOC_VENDOR_UNKNOWN;
 }
 

src/riscv/uarch.c

@@ -4,6 +4,7 @@
 #include <string.h>
 
 #include "uarch.h"
+#include "udev.h"
 #include "../common/global.h"
 
 typedef uint32_t MICROARCH;
@@ -12,6 +13,7 @@ struct uarch {
   MICROARCH uarch;
   char* uarch_str;
   char* cpuinfo_str;
+  struct riscv_cpuinfo* ci;
 };
 
 enum {
@@ -21,13 +23,20 @@ enum {
   UARCH_U74,
   // THEAD
   UARCH_C906,
-  UARCH_C910
+  UARCH_C910,
+  // SPACEMIT
+  UARCH_X60
 };
 
 #define UARCH_START if (false) {}
 #define CHECK_UARCH(arch, cpu, cpuinfo_str, uarch_str, str, uarch, vendor) \
    else if (strcmp(cpuinfo_str, uarch_str) == 0) fill_uarch(arch, cpu, str, uarch, vendor);
-#define UARCH_END else { printBug("Unknown microarchitecture detected: uarch='%s'", cpuinfo_str); fill_uarch(arch, cpu, "Unknown", UARCH_UNKNOWN, CPU_VENDOR_UNKNOWN); }
+#define UARCH_END else { printWarn("Unknown microarchitecture detected: uarch='%s'", cpuinfo_str); fill_uarch(arch, cpu, "Unknown", UARCH_UNKNOWN, CPU_VENDOR_UNKNOWN); }
+
+#define ARCHID_START if (false) {}
+#define CHECK_ARCHID(arch, marchid_val, str, uarch, vendor) \
+  else if (arch->ci->marchid == (unsigned long) marchid_val) fill_uarch(arch, cpu, str, uarch, vendor);
+#define ARCHID_END else { printWarn("Unknown microarchitecture detected: marchid=0x%.8X", arch->ci->marchid); fill_uarch(arch, cpu, "Unknown", UARCH_UNKNOWN, CPU_VENDOR_UNKNOWN); }
 
 void fill_uarch(struct uarch* arch, struct cpuInfo* cpu, char* str, MICROARCH u, VENDOR vendor) {
   arch->uarch = u;
@@ -39,14 +48,8 @@ void fill_uarch(struct uarch* arch, struct cpuInfo* cpu, char* str, MICROARCH u,
 // https://elixir.bootlin.com/linux/latest/source/Documentation/devicetree/bindings/riscv/cpus.yaml
 // SiFive: https://www.sifive.com/risc-v-core-ip
 // T-Head: https://www.t-head.cn/product/c906
-struct uarch* get_uarch_from_cpuinfo_str(char* cpuinfo_str, struct cpuInfo* cpu) {
-  struct uarch* arch = emalloc(sizeof(struct uarch));
+struct uarch* get_uarch_from_cpuinfo_str(char* cpuinfo_str, struct cpuInfo* cpu, struct uarch* arch) {
   arch->cpuinfo_str = cpuinfo_str;
-  if(cpuinfo_str == NULL) {
-    printWarn("get_uarch_from_cpuinfo: Unable to detect microarchitecture, cpuinfo_str is NULL");
-    fill_uarch(arch, cpu, "Unknown", UARCH_UNKNOWN, CPU_VENDOR_UNKNOWN);
-    return arch;
-  }
 
   // U74/U74-MC:
   // SiFive says that U74-MC is "Multicore: four U74 cores and one S76 core" while
@@ -70,6 +73,41 @@ struct uarch* get_uarch_from_cpuinfo_str(char* cpuinfo_str, struct cpuInfo* cpu)
   return arch;
 }
 
+// Use marchid to get the microarchitecture
+struct uarch* get_uarch_from_riscv_cpuinfo(struct cpuInfo* cpu, struct uarch* arch) {
+
+  ARCHID_START
+  CHECK_ARCHID(arch, 0x8000000058000001, "X60", UARCH_X60, CPU_VENDOR_SPACEMIT) // https://github.com/Dr-Noob/cpufetch/issues/286
+  ARCHID_END
+
+  return arch;
+}
+
+struct uarch* get_uarch(struct cpuInfo* cpu) {
+  char* cpuinfo_str = get_uarch_from_cpuinfo();
+  struct uarch* arch = emalloc(sizeof(struct uarch));
+  arch->uarch = UARCH_UNKNOWN;
+  arch->ci = NULL;
+
+  if (cpuinfo_str == NULL) {
+    printWarn("get_uarch_from_cpuinfo: Unable to detect microarchitecture using uarch: cpuinfo_str is NULL");
+    arch->ci = get_riscv_cpuinfo();
+
+    if (arch->ci == NULL || arch->ci->marchid == 0)
+      printWarn("get_riscv_cpuinfo: Unable to get marchid from udev");
+    else
+      arch = get_uarch_from_riscv_cpuinfo(cpu, arch);
+  }
+  else {
+    arch = get_uarch_from_cpuinfo_str(cpuinfo_str, cpu, arch);
+  }
+
+  if (arch->uarch == UARCH_UNKNOWN)
+    fill_uarch(arch, cpu, "Unknown", UARCH_UNKNOWN, CPU_VENDOR_UNKNOWN);
+
+  return arch;
+}
+
 char* get_str_uarch(struct cpuInfo* cpu) {
   return cpu->arch->uarch_str;
 }

src/riscv/uarch.h

@@ -9,6 +9,6 @@ struct uarch;
 char* get_arch_cpuinfo_str(struct cpuInfo* cpu);
 char* get_str_uarch(struct cpuInfo* cpu);
 void free_uarch_struct(struct uarch* arch);
-struct uarch* get_uarch_from_cpuinfo_str(char* cpuinfo_str, struct cpuInfo* cpu);
+struct uarch* get_uarch(struct cpuInfo* cpu);
 
 #endif

src/riscv/udev.c

@@ -7,6 +7,9 @@
 #define _PATH_DEVTREE          "/proc/device-tree/compatible"
 #define CPUINFO_UARCH_STR      "uarch\t\t: "
 #define CPUINFO_EXTENSIONS_STR "isa\t\t: "
+#define CPUINFO_RISCV_MVENDORID "mvendorid\t:"
+#define CPUINFO_RISCV_MARCHID   "marchid\t\t:"
+#define CPUINFO_RISCV_MIMPID    "mimpid\t\t:"
 #define DEVTREE_HARDWARE_FIELD 0
 
 char* get_field_from_devtree(int DEVTREE_FIELD) {
@@ -75,6 +78,52 @@ char* parse_cpuinfo_field(char* field_str) {
   return ret;
 }
 
+unsigned long parse_cpuinfo_field_uint64(char* field_str) {  
+  int filelen;
+  char* buf;
+  if((buf = read_file(_PATH_CPUINFO, &filelen)) == NULL) {
+    printWarn("read_file: %s: %s", _PATH_CPUINFO, strerror(errno));
+    return 0;
+  }
+
+  char* tmp = strstr(buf, field_str);
+  if(tmp == NULL) return 0;
+  tmp += strlen(field_str);
+
+  char* end;
+  errno = 0;
+  unsigned long ret = strtoul(tmp, &end, 16);
+  if (errno != 0) {
+    printWarn("strtoul: %s: %s", strerror(errno), tmp);
+    return 0;
+  }
+  
+  return ret;
+}
+
+// Creates and fills in the riscv_cpuinfo struct (which contains
+// mvendorid, marchid and mimpid) using cpuinfo to fetch the values.
+// 
+// Every RISC-V hart (hardware thread) [1] provides a
+// marchid (Machine Architecture ID register) CSR that encodes its
+// base microarchitecture [2]. For more information about
+// marchid and the rest of values, see [3].
+// [1] https://groups.google.com/a/groups.riscv.org/g/sw-dev/c/QKjUDjz_vKo
+// [2] https://github.com/riscv/riscv-isa-manual/blob/main/marchid.md
+// [3] https://five-embeddev.com/riscv-priv-isa-manual/Priv-v1.12/machine.html#machine-architecture-id-register-marchid
+struct riscv_cpuinfo *get_riscv_cpuinfo(void) {
+  struct riscv_cpuinfo* ci = emalloc(sizeof(struct riscv_cpuinfo));
+
+  ci->mvendorid = parse_cpuinfo_field_uint64(CPUINFO_RISCV_MVENDORID);
+  ci->marchid = parse_cpuinfo_field_uint64(CPUINFO_RISCV_MARCHID);
+  ci->mimpid = parse_cpuinfo_field_uint64(CPUINFO_RISCV_MIMPID);
+
+  if (ci->mvendorid == 0 && ci->mvendorid == 0 && ci->mvendorid == 0)
+    return NULL;
+
+  return ci;
+}
+
 char* get_hardware_from_devtree(void) {
   return get_field_from_devtree(DEVTREE_HARDWARE_FIELD);
 }

src/riscv/udev.h

@@ -5,8 +5,16 @@
 
 #define UNKNOWN -1
 
+// https://elixir.bootlin.com/linux/v6.10.6/source/arch/riscv/include/asm/cpufeature.h#L21
+struct riscv_cpuinfo {
+  unsigned long mvendorid;
+  unsigned long marchid;
+  unsigned long mimpid;
+};
+
 char* get_hardware_from_devtree(void);
 char* get_uarch_from_cpuinfo(void);
 char* get_extensions_from_cpuinfo(void);
+struct riscv_cpuinfo *get_riscv_cpuinfo(void);
 
 #endif