[PATCH] amd/lib: refactor function is_reset_enable

Michael Strawbridge michael.strawbridge at amd.com
Tue Jan 28 20:43:54 UTC 2025


Hi Vitaly,

One small thing below.

On 1/28/25 14:24, vitaly.prosyak at amd.com wrote:
> From: Vitaly Prosyak <vitaly.prosyak at amd.com>
> 
> This function takes a reset type mask (`AMDGPU_RESET_TYPE_FULL`,
> `AMDGPU_RESET_TYPE_SOFT_RESET`, AMDGPU_RESET_TYPE_PER_PIPE
> or `AMDGPU_RESET_TYPE_PER_QUEUE`) and attempts to open and
> read the corresponding sysfs file, e.g.,
> `/sys/bus/pci/devices/0000:08:00.0/gfx_reset_mask`.
> It then iterates through predefined keywords such as `full`,
> `soft`, and `queue`, validating whether AMDGPU exposes the
> requested operation. If the requested operation is not available,
> the test is skipped.
> 
> The previous implementation included redundant steps and failed
> in certain cases. This refactored version simplifies the logic,
> improves reliability, and ensures consistent behavior across various
> scenarios.
> 
> Cc: Christian Koenig <christian.koenig at amd.com>
> Cc: Alexander Deucher <alexander.deucher at amd.com>
> Cc: Jesse Zhang <jesse.zhang at amd.com>
> Cc: Michael Strawbridge <michael.strawbridge at amd.com>
> 
> Signed-off-by: Vitaly Prosyak <vitaly.prosyak at amd.com>
> ---
>  lib/amdgpu/amd_ip_blocks.c | 79 +++++++++++++++++++++++++-------------
>  1 file changed, 53 insertions(+), 26 deletions(-)
> 
> diff --git a/lib/amdgpu/amd_ip_blocks.c b/lib/amdgpu/amd_ip_blocks.c
> index 2b1b84414..f063ad0ab 100644
> --- a/lib/amdgpu/amd_ip_blocks.c
> +++ b/lib/amdgpu/amd_ip_blocks.c
> @@ -1006,11 +1006,27 @@ asic_rings_readness(amdgpu_device_handle device_handle, uint32_t mask,
>  bool
>  is_reset_enable(enum amd_ip_block_type ip_type, uint32_t reset_type, const struct pci_addr *pci)
>  {
> -	char cmd[256];
> -	FILE *fp, *fp2;
> -	char buffer[100];
> +	const struct reset_arr {
> +		const char *name;
> +		unsigned int reset;
> +	} reset_arr[] = {
> +		{"full",	AMDGPU_RESET_TYPE_FULL		},
> +		{"soft",	AMDGPU_RESET_TYPE_SOFT_RESET},
> +		{"queue",	AMDGPU_RESET_TYPE_PER_QUEUE },
> +		{"pipe",	AMDGPU_RESET_TYPE_PER_PIPE },
> +		{NULL, 0}
> +	};
> +
>  	bool enable = false;
> -	char reset_mask[100];
> +	char cmd[256];
> +	FILE *fp;
> +	char buffer[128];
> +
> +	char *token, *buf;
> +	char reset_mask[32];
> +	uint32_t mask = 0;
> +	int found = 0;
> +	int i;
>  
>  	if (ip_type == AMD_IP_GFX)
>  		snprintf(reset_mask, sizeof(reset_mask) - 1, "gfx_reset_mask");
> @@ -1019,33 +1035,44 @@ is_reset_enable(enum amd_ip_block_type ip_type, uint32_t reset_type, const struc
>  	else
>  		snprintf(reset_mask, sizeof(reset_mask) - 1, "sdma_reset_mask");
>  
> -	if( pci)
> -		snprintf(cmd, sizeof(cmd) - 1, "sudo cat /sys/kernel/debug/dri/%04x:%02x:%02x.%01x/name |grep -oP '(?<=dev=)[0-9:.]+'",
> -			pci->domain, pci->bus, pci->device, pci->function);
> -	else
> -		snprintf(cmd, sizeof(cmd) - 1, "sudo cat /sys/kernel/debug/dri/0/name |grep -oP '(?<=dev=)[0-9:.]+'");
> +	snprintf(cmd, sizeof(cmd) - 1, "sudo cat /sys/bus/pci/devices/%04x:%02x:%02x.%01x/%s",
> +			pci->domain, pci->bus, pci->device, pci->function, reset_mask);
>  
>  	fp = popen(cmd, "r");
> -	if (fp == NULL)
> +	if (fp == NULL) {
> +		igt_kmsg("***FAILURE popen %s LINE %d FILE %s\n", cmd, __LINE__, __FILE__);
>  		return false;
> +	}
>  
> -	if (fgets(buffer, 13, fp) != NULL) {
> -		snprintf(cmd, sizeof(cmd) - 1, "sudo cat /sys/bus/pci/devices/%s/%s | grep -oP '%s'",
> -			buffer, reset_mask,
> -			reset_type & AMDGPU_RESET_TYPE_FULL ? "full" :
> -			reset_type & AMDGPU_RESET_TYPE_SOFT_RESET ? "soft" :
> -			reset_type & AMDGPU_RESET_TYPE_PER_QUEUE ? "queue" : "pipe");
> -
> -		fp2 = popen(cmd, "r");
> -		if (fp2 == NULL) {
> -			pclose(fp);
> -			return false;
> +	buf = fgets(buffer, sizeof(buffer)-1, fp);
> +
> +	//igt_kmsg("cmd: %s buf: %s LINE %d FILE %s\n",
> +	//		cmd, buf != NULL ? buf : "NULL", __LINE__, __FILE__);

Did you mean to keep the 2 above commented lines?

Besides this you can add:
Reviewed-by: Michael Strawbridge <michael.strawbridge at amd.com>

> +
> +	if (buf != NULL) {
> +		token = strtok(buf, " \n");
> +		while (token != NULL) {
> +			for (i = 0; reset_arr[i].name != NULL; i++) {
> +				if (reset_type == reset_arr[i].reset &&
> +					strcmp(token, reset_arr[i].name) == 0) {
> +					mask |= reset_arr[i].reset;
> +					found = 1;
> +					break;
> +				}
> +			}
> +			if (!found)
> +				token = strtok(NULL, " \n");
>  		}
> -
> -		if (fgets(buffer, 13, fp2) != NULL)
> -			enable = true;
> -		pclose(fp2);
> +	} else {
> +		igt_kmsg("***FAILURE fgets %s LINE %d FILE %s\n",
> +				buffer, __LINE__, __FILE__);
>  	}
> +	if (mask == reset_type)
> +		enable = true;
> +	else
> +		igt_kmsg("***FAILURE mask found 0x%x requested 0x%x operation is not supported LINE %d FILE %s\n",
> +				mask, reset_type, __LINE__, __FILE__);
> +
>  	pclose(fp);
>  
>  	return enable;
> @@ -1105,7 +1132,7 @@ int get_pci_addr_from_fd(int fd, struct pci_addr *pci)
>  
>  	if (ret != 4) {
>  		igt_info("error %s Unable to extract PCI device address from '%s\n",
> -				__FUNCTION__, buf);
> +				__func__, buf);
>  		return -ENOENT;
>  	}
>  



More information about the igt-dev mailing list